mesh/merge-gate head e7ff7c3045fc merged into trunk afb94d925e94 — pass: builds mesh-provisioner-client → ace, g14, novox, shanks; no bus step; every machi…
mesh/repo-check head e7ff7c3045fc merged into trunk afb94d925e94 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/merge-gate head 11761637cf81 merged into trunk e47d0694c752 — pass: builds gitea → novox; no bus step; every machine composes with the change as it di…
mesh/repo-check head 11761637cf81 merged into trunk e47d0694c752 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/merge-gate head 0d3f0fd3fc89 merged into trunk 7f77ce663a9e — pass: builds gitea → novox; no bus step; every machine composes with the change as it di…
mesh/repo-check head 0d3f0fd3fc89 merged into trunk 7f77ce663a9e — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/merge-gate head 695446729264 merged into trunk 580fadb63ecb — pass: builds mesh-delivery → novox; no bus step; every machine composes with the change …
mesh/repo-check head 695446729264 merged into trunk 580fadb63ecb — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/merge-gate head 5f002441b7f0 merged into trunk 580fadb63ecb — pass: builds mesh-delivery → novox; no bus step; every machine composes with the change …
mesh/repo-check head 5f002441b7f0 merged into trunk 580fadb63ecb — fail: its merge-check.sh failed: --- FAIL: TestAGroupOfTwoWithGreenChecksGetsOneAskAndOn…
mesh/merge-gate head 30d3bb94667f merged into trunk e40ec47058e6 — pass: builds mesh-delivery → novox; no bus step; every machine composes with the change …
mesh/repo-check head 30d3bb94667f merged into trunk e40ec47058e6 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/merge-gate head bc1acddaa4e5 merged into trunk 798d6fd68589 — pass: builds mesh-delivery → novox; no bus step; every machine composes with the change …
mesh/repo-check head bc1acddaa4e5 merged into trunk 798d6fd68589 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/merge-gate head b0ec4f6f1405 merged into trunk d3dd46a9fdb4 — pass: builds mesh-delivery → novox; no bus step; every machine composes with the change …
mesh/repo-check head b0ec4f6f1405 merged into trunk d3dd46a9fdb4 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/repo-check head 3bc4215d872c merged into trunk 44e7002d3c84 beside mesh-controller#378@0149a3d7 — pass: its merge-check.sh passed, each part in its to…
mesh/delivery stopped: its walk plan-1792289060772890935 was stopped by a person (mesh-controller): the operator, via telegram (user id verified), chose …
mesh/repo-check head 68e3326d4cbc merged into trunk 44e7002d3c84 beside mesh-controller#378@b56c1d00 — pass: its merge-check.sh passed, each part in its to…
mesh/merge-gate head 68e3326d4cbc merged into trunk 44e7002d3c84 — pass: group feat/chain-on-the-bus composed: every machine composes with the change as it…
mesh/merge-gate head 7383bc31703e merged into trunk 0c56f603059b — pass: builds gitea, mesh-delivery → novox; no bus step; every machine composes with the …
mesh/repo-check head 7383bc31703e merged into trunk 0c56f603059b — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/merge-gate head 684e8801c420 merged into trunk 3395af96bc34 — pass: builds mesh-knowledge → novox; no bus step; every machine composes with the change…
mesh/repo-check head 684e8801c420 merged into trunk 3395af96bc34 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
WithdrawRecord only ended the waiting ask locally; the router kept it open, so
the operator still saw Approve/Decline for a withdrawn record and an answer to
it was dropped silently. It now also publishes a cancel on the router's
subject once the withdraw is saved, as the controller's asker does; a cancel
the router cannot be told of is logged and the withdraw still stands. The
module's existing "uses": ["operator-channel"] already grants this; no manifest
change needed, since the ask and cancel subjects are both under that seat.
A dropped proposal or a stale reservation stayed proposed forever: there was
no way to say the operator decided against it. knowledge-base.withdraw takes
a reserved or proposed record to withdrawn as a new version naming why, ends
a waiting operator's ask and clears a pending review in the same version,
and refuses an already accepted, superseded or withdrawn record. The number
stays spent either way.
The old Withdraw (the background job for reservations never proposed) is
renamed WithdrawStale to make room for the new, caller-driven WithdrawRecord.
mesh/repo-check head 4d03755c6598 merged into trunk ad80048c22d8 beside mesh-controller#375@91a7c1ea — pass: its merge-check.sh passed, each part in its to…
mesh/merge-gate head 4d03755c6598 merged into trunk ad80048c22d8 — pass: group feat/backup-before-send composed: every machine composes with the change as …
mesh/merge-gate head bf3b3a5e362b merged into trunk f076cadecedc — pass: builds mesh-issues, mesh-knowledge → novox; no bus step; every machine composes wi…
mesh/repo-check head bf3b3a5e362b merged into trunk f076cadecedc — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/repo-check head c898a937a1e3 merged into trunk f076cadecedc beside mesh-controller#375@a39d6ce7 — pass: its merge-check.sh passed, each part in its to…
mesh/merge-gate head 0771797431ed merged into trunk f69f84b9e54c beside mesh-controller#374@2d24fbb7 — pass: builds gitea, mesh-delivery → novox; no bus st…
mesh/repo-check head 0771797431ed merged into trunk f69f84b9e54c beside mesh-controller#374@2d24fbb7 — pass: its merge-check.sh passed, each part in its to…
mesh/repo-check head 86d5706182fd merged into trunk f69f84b9e54c — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
A group has no order: mesh-delivery asks the controller for none, reads no after:
line (it says once that the line is not read), merges every member in one pass
once the composed check passed and the Approve covers the heads, and gives the
plan its word once every member is merged. A failed member stops no other. The
composed verdict sets each member's mesh/merge-gate (pending, saying why, until
one stands), the groups answer says composed_gate, and a trunk move asks the
out-of-date members' checks at once and merges on the same Approve (issue 1047).
The review finishes before the push and its record is a Reviewed-by line in the
description, so the mark repeated what the push meant. mesh-delivery asks the
operator when a pull request's checks pass, keeps an ask through a trunk move
by asking the check again at once, and hands the merge tool the approved heads,
which replace the forge review of each sibling.
mesh/merge-gate head 7104df6fa284 merged into trunk 5e6f5b8f6d99 — pass, with a note: builds artifact-store-tools, asus-zephyrus-g14, audit-logger, baserow…
mesh/repo-check head 7104df6fa284 merged into trunk 5e6f5b8f6d99 — pass: THE CHANGE ALTERS ITS OWN CHECK (merge-check.sh): main's version judged it; the ch…
mesh-catalog #350 removed both modules from the catalogue; this change's edits to them go with them, and mesh-issues keeps both #350's change and its `unknown`. The branch's tree is the reviewed tree merged with main 5e6f5b8f6d.
mesh/merge-gate head 629aafdea205 merged into trunk 54b3ce503e79 — pass: builds mesh-issues, new: modules/build-agent, modules/route-proxy → novox; no bus …
mesh/repo-check head 629aafdea205 merged into trunk 54b3ce503e79 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/delivery covered after its stop: a later walk that contains it was delivered: plan-1792260860441586760 (novox/mesh-catalog at 5e7567e5 into main, do…
mesh-controller #370 merged and registered route-proxy from mesh-controller; its plan then built
route-proxy from the catalogue, which still held it, and failed. This member removes it there.
mesh/delivery-group group feat/whole-modules delivering: 1 of 2 delivered; novox/mesh-catalog@795806efa7ad, stopped for a member delivered since, waits for a l…
mesh-delivery reads a pull request's after: lines only when it records a new head, so the
line added to #350's description takes effect with this head.
mesh/merge-gate checking again: the gate refused `unknown` because the controller judged with a node-engine validator pinned before mesh-host #117; mesh-co…
mesh/repo-check checking again: the gate refused `unknown` because the controller judged with a node-engine validator pinned before mesh-host #117; mesh-co…
Every directory and archive resource carries `unknown`: move-out for a home folder whose program writes nothing there (the node-engine, delivered in mesh-host #117, moves the rest to ~/.local/state/mesh-moved/<module>/ with a list), module for ~/.ssh and ~/.claude, program elsewhere; `knows` keeps what a program or the operator relies on (powerlevel10k's compiled files, i3's scripts and the asus/g14 drop-ins, dunst's slack rule, betterbird's records, node-env's forge credential, issue 1033). Two shared folders are no longer declared. Each module's README says it. checks/owned holds every manifest to it; checks/sshdir stays while main's merge check runs it (issue 1035). Tests that pin a manifest follow the new key only. Reviewed in three rounds.
mesh/delivery-group group feat/whole-modules rejected: its order contradicts itself: novox/mesh-catalog@c47d94925811 and novox/mesh-controller@48f5a7d3a99a: bu…
mesh/merge-gate checking again: issue 1043: mesh-controller #370's new head 0125ccd7 lets a member's own gate see the move through the clones beside it; th…
mesh/repo-check checking again: issue 1043: mesh-controller #370's new head 0125ccd7 lets a member's own gate see the move through the clones beside it; th…
mesh/merge-gate head 40eed3a036db merged into trunk 0d2fc5e4d79c — pass: builds mesh-delivery → novox; no bus step; every machine composes with the change …
mesh/repo-check head 40eed3a036db merged into trunk 0d2fc5e4d79c — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/merge-gate head 05e6b583e2e3 merged into trunk ce56b491cffd — pass: builds ssh-client → ace, g14, novox, shanks; no bus step; every machine composes w…
mesh/repo-check head 05e6b583e2e3 merged into trunk ce56b491cffd — pass: THE CHANGE ALTERS ITS OWN CHECK (merge-check.sh): main's version judged it; the ch…
The sweep from issue 1017 moved known_hosts2, rc and environment, which ssh
and sshd read by default, and moved a link to a folder of keys, breaking every
login through it. It could move a file a tool was still writing. When ~/.local
was on another file system, every sweep failed at its first move and left an
empty folder each round, seen only in the journal. Its list was written after
the moves with errors ignored.
Now those files and linked key folders stay; a link whose target, or the folder
it names, cannot be checked is left and named; a copy of authorized_keys is
moved at any depth; a path changed in the last minute, at any depth, waits for
the next sweep; a retired folder on another file system fails the sweep before
anything is made; the list names each move before it is made; and each sweep's
outcome is kept for ssh_client_check, which shows a failed sweep as a problem.
A key's first line is read only so far.
The catalogue's new checks/sshdir refuses any manifest but ssh-client naming a
path under ~/.ssh (ADR 0415 §6). The README cites §7 for config.d.
mesh/repo-check head 94f65cccd9c1 merged into trunk a7d526838ab0 beside mesh-controller#368@0f21fe78 — pass: its merge-check.sh passed, each part in its to…
mesh/merge-gate head acf075fa63cc merged into trunk 9675babf528b — pass: builds ssh-client → ace, g14, novox, shanks; no bus step; every machine composes w…
mesh/repo-check head acf075fa63cc merged into trunk 9675babf528b — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
ssh-client owns ~/.ssh, but treated every file it did not place as found, so a
predecessor's drop-in and old copies stayed there for weeks. Its process now
sweeps at start and every ten minutes: keys at any depth (and links to them),
authorized_keys, authorized_keys2 and sockets stay, and a folder holding one is
gone into; everything else moves to ~/.local/state/ssh-client/retired/<UTC
time>/, a folder never shared with another sweep, with a list beside it, which
ssh_client_check shows. Nothing is deleted.
mesh/delivery-group group feat/events-and-queues rejected: its order contradicts itself: novox/mesh-catalog@52f70f42b749 and novox/mesh-controller@0f21fe7888be…
mesh/repo-check head 52f70f42b749 merged into trunk a7d526838ab0 beside mesh-controller#368@0f21fe78 — pass: its merge-check.sh passed, each part in its to…
mesh/merge-gate head 3a2560dc51fc merged into trunk 82df3bd58913 — pass: builds gitea → novox; no bus step; every machine composes with the change as it di…
mesh/repo-check head 3a2560dc51fc merged into trunk 82df3bd58913 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/merge-gate head 661b83249c2a merged into trunk bb291b5ad8af — pass: builds gitea, ssh-client → ace, g14, novox, shanks; no bus step; every machine com…
mesh/repo-check head 661b83249c2a merged into trunk bb291b5ad8af — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/merge-gate head b159efd65673 merged into trunk 522780d5fcfa — pass: builds gitea → novox; no bus step; every machine composes with the change as it di…
mesh/repo-check head b159efd65673 merged into trunk 522780d5fcfa — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/merge-gate head 065ecd85f53e merged into trunk 734a77f5a91a — pass: builds gitea → novox; no bus step; every machine composes with the change as it di…
mesh/repo-check head 065ecd85f53e merged into trunk 734a77f5a91a — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/merge-gate head ae9cda8531b3 merged into trunk 017b729bc95d — pass: builds gitea, ssh-client → ace, g14, novox, shanks; no bus step; every machine com…
mesh/repo-check head ae9cda8531b3 merged into trunk 017b729bc95d — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/delivery covered after its failure: a later walk that contains it was delivered: plan-1792195559895153198 (novox/mesh-catalog at 82df3bd5 into main,…
mesh/merge-gate head 6f45570d4c23 merged into trunk 017b729bc95d — fail: builds gitea, ssh-client → ace, g14, novox, shanks; no bus step; 4 problem(s) the …
mesh/repo-check head 6f45570d4c23 merged into trunk 017b729bc95d — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/merge-gate head deb2020bd47d merged into trunk 5652a1494493 — pass: builds postgres, sent nowhere; no bus step; 2 wait(s) for a person; every machine …
mesh/repo-check head deb2020bd47d merged into trunk 5652a1494493 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/repo-check head 9acd1f51c5cf merged into trunk 8d012fe4e270 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
The client decodes a plan strictly, and since the controller carries these fields
(issues 923 and 928) it refused the plans of gitea, keycloak and claude-code.
mesh/repo-check head d2992f81d17f merged into trunk 3817099b2ca2 beside mesh-controller#363@df5f01bf — pass: its merge-check.sh passed, each part in its to…
postgres, keycloak, mesh-provisioner-client, route-proxy and route-adapter now
build against the feature's mesh-sdk: the harness package, the vault-name key
and the helper's fixes.
mesh/merge-gate head 6005c6001e2f merged into trunk 832a0afd1814 — pass: builds keycloak, postgres, sent nowhere; no bus step; 3 wait(s) for a person; ever…
mesh/repo-check head 6005c6001e2f merged into trunk 832a0afd1814 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
postgres and keycloak start their moved path beside Serve, and a watch
asked before Serve runs is refused by the runtime. Run returned on that
refusal and main only logged it, so a start that lost the race left the
provider answering nobody and retiring nothing until its next restart:
postgres on novox at 16:52, 18:29 and 18:48, on ace at 16:22, which kept
issue 937's retirement condition on novox from clearing. Run now asks
again, the wait doubling from a second to a minute, until the runtime
answers or its context ends.
mesh/merge-gate head d7a880a60878 merged into trunk 3ee754eb3b0c — pass: builds keycloak, postgres, sent nowhere; no bus step; 3 wait(s) for a person; ever…
mesh/repo-check head d7a880a60878 merged into trunk 3ee754eb3b0c — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
The controller clears a provider's retire-waiting condition only on its
settled word, and a provider says settled only for a set its own memory
held. A provider restarted after the set stopped waiting (here: the
delivery of the grant-file fix) never says it, so the condition kept
asking the operator to approve retiring what nobody waits on. The first
pass of a process that finds every consumer asked now says settled once,
with no set; the controller clears both conditions on it.
mesh/merge-gate head 41cdd032084e merged into trunk 6268f8f94bda — pass: builds keycloak, postgres, sent nowhere; no bus step; 3 wait(s) for a person; ever…
mesh/repo-check head 41cdd032084e merged into trunk 6268f8f94bda — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
Only a moved consumer's plan names postgres, so on the moved path every
consumer still given by the controller's grant file was counted unasked
and offered for retirement (ace 5 of 5, novox 7 of 9), and could have
been retired unasked where the bound allowed. The moved path now reads
MESH_RECEIVES again: its logins are asked, the ones no plan names are
made and checked by the harness's grant pass until R1 (issue 682), and
an unreadable grant file retires nothing (issue 241). The harness's
grant pass becomes apply, identically in keycloak's copy.
mesh/merge-gate head 77d751fe830d merged into trunk 08188663faff — pass: builds mesh-knowledge, messenger → novox; no bus step; every machine composes with…
mesh/repo-check head 77d751fe830d merged into trunk 08188663faff — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
The router refused such an ask with 'no channel', so the asker waited for an Approve nobody saw. It is now held as an open ask with the per-kind reason, judged again when the kinds or a channel's standing change and every minute, and the knowledge base says held and why.
mesh/repo-check head d4a9260847ec merged into trunk 8b53a3663244 beside mesh-controller#360@f4ddddd5 — pass: its merge-check.sh passed, each part in its to…
mesh/delivery-group group feat/900-node-files checking: composed check out of date: novox/mesh-controller@210f7cf5e64b was judged against trunk 611e2275cf7c, t…
mesh/merge-gate head 2d1b20364f9b merged into trunk 3db85a0dde83 beside mesh-controller#357@114c99e0 — pass: builds mesh-delivery → novox; no bus step; eve…
mesh/repo-check head 2d1b20364f9b merged into trunk 3db85a0dde83 beside mesh-controller#357@114c99e0 — pass: its merge-check.sh passed, each part in its to…
mesh/merge-gate head a31ad48b024d merged into trunk ee1b15d7f066 — pass, with a note: builds adwaita, artifact-store-tools, asus-zephyrus-g14, baserow, bet…
mesh/repo-check head a31ad48b024d merged into trunk ee1b15d7f066 — pass: THE CHANGE ALTERS ITS OWN CHECK (merge-check.sh): main's version judged it; the ch…
mesh/merge-gate head 8efe499db965 merged into trunk 173c5d5ef434 — pass: builds mesh-delivery → novox; no bus step; every machine composes with the change …
mesh/repo-check head 8efe499db965 merged into trunk 173c5d5ef434 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/merge-gate head c7c784161dcd merged into trunk 0fdacb98399d — pass: builds mesh-delivery → novox; no bus step; every machine composes with the change …
mesh/repo-check head c7c784161dcd merged into trunk 0fdacb98399d — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
After mesh-host#114's recheck passed, group 57-config-and-provisioning kept its failed composed verdict for the same heads and was never asked again. A recheck now marks the member's group's composed check out of date, answered or still running, so it is asked again once every member is judged again and ready; a late answer from the old run is dropped.
Co-Authored-By: Claude <noreply@anthropic.com>
mesh/merge-gate head b0e4a12c8840 merged into trunk 21a499ba5976 — pass: builds mesh-mail → novox; no bus step; every machine composes with the change as i…
mesh/repo-check head b0e4a12c8840 merged into trunk 21a499ba5976 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
Review round 1: the test judged Headline+Explanation+Details, but messenger refuses an ask as a whole on
Headline, Explanation, OnExpiry and each option's Label and Does alone (asks.go's TakeAsk); Details is
filtered line by line instead (holder.go's sayOn). send.go's Ask() now builds that exact judged text and
checks it with messenger's real content rule, carried at run time as content_rule.go (messenger's
content.go, renamed, byte-identical, pinned by copies_test.go) rather than only in a test copy. When a
subject or first line the masking does not catch (an IP, a hardware address, a path-shaped reference, a
hex hash, ...) would still refuse the whole ask, the Explanation drops the subject and first lines and
points at Details instead, so the ask always reaches the operator. A recipient written with a display
name ('Name <addr>') is now matched against the operator's own accounts, and masked, by its address alone
(net/mail.ParseAddress).
messenger's content rule (ADR 0253 §3) refuses any message carrying a mail address, so mail_send's ask
never reached the operator: no send could ever be approved. A recipient that is one of the operator's own
mesh-mail accounts is now named by that account's name; any other loses its address and host; the subject
and body's first lines have every address and link masked before they reach the ask. The mail itself is
sent unchanged.
Tests run messenger's real content rule against the asks mail_send builds (content.go copied verbatim as
content_test.go, pinned byte-identical by copies_test.go, since messenger's package main cannot be
imported from another module).
mesh/merge-gate head de9d29dc657d merged into trunk 62ec68355cd1 — pass: builds gitea, mesh-delivery → novox; no bus step; every machine composes with the …
mesh/repo-check head de9d29dc657d merged into trunk 62ec68355cd1 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh-delivery's groups names each member's pull request, head, base and whether it is open (member_details), so the controller can read a member's siblings beside its check. The forge's holder keeps the siblings each verdict read, puts that verdict out of date when a sibling gets a new head, closes, or its base moves by anything but its own merge, and the merge tool merges such a member only when mesh/delivery-group passed at the heads read and each sibling head is approved. The statuses name the siblings, or say they could not be learned.
Co-Authored-By: Claude <noreply@anthropic.com>
mesh/merge-gate head 3f97d2d50d57 merged into trunk 6a5e6510cab2 — pass: builds mesh-mail-notifier → g14, shanks; no bus step; every machine composes with …
mesh/repo-check head 3f97d2d50d57 merged into trunk 6a5e6510cab2 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/merge-gate head 6112233649d3 merged into trunk 8c2d7e5f97f1 — pass: builds clipmenu → g14, shanks; no bus step; every machine composes with the change…
mesh/repo-check head 6112233649d3 merged into trunk 8c2d7e5f97f1 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/merge-gate head 21ebbffed469 merged into trunk 8c2d7e5f97f1 — pass: builds clipmenu → g14, shanks; no bus step; every machine composes with the change…
mesh/repo-check head 21ebbffed469 merged into trunk 8c2d7e5f97f1 — fail: its merge-check.sh failed: --- FAIL: TestTheLauncherWaitsQuietlyWhileAnotherClipme…
mesh/merge-gate head 31120ed58029 merged into trunk c9848f4fb85c — pass: builds mesh-mail-notifier → g14, shanks; no bus step; every machine composes with …
mesh/repo-check head 31120ed58029 merged into trunk c9848f4fb85c — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
A subscribe the runtime refuses no longer calls os.Exit(1), which took mail-notifier.status and
mail_notifier_health down with it: watch now retries with backoff (2s doubling to 1 minute), logging each
refusal, as mesh-knowledge's hear does. Also: the run() test proves liveness by having the watcher answer a
ping after run returns, instead of checking a channel only the test's own deferred cleanup could close; and
the manifest test is renamed to say what it checks (no process resource at all).
The notify process's stdin is empty under systemd, so stdio.Serve returned at EOF, main returned, and
systemd restarted it every five seconds: the watch loop never ran long enough to show a notice. The watch
loop now starts beside stdio.Serve in the one process the node's runtime launches and keeps alive, as
mesh-knowledge's serve does. No resource declares a separate notify process any more.
mesh/merge-gate head c070bea782aa merged into trunk 7443938fef69 — pass: builds clipmenu → g14, shanks; no bus step; every machine composes with the change…
mesh/repo-check head c070bea782aa merged into trunk 7443938fef69 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/merge-gate head c816cafc00bd merged into trunk de6657c846b9 — pass: builds kernel → ace, g14, novox, shanks; no bus step; every machine composes with …
mesh/repo-check head c816cafc00bd merged into trunk de6657c846b9 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
clipmenud is declared as the operator's user unit, enabled and running, with a launcher that finds the display. Copy reads the clipboard back and fails with the reason. History and paste answer counts, times and lengths, never content.
mesh/merge-gate head d4b8b2470159 merged into trunk aee97bf8ad42 — pass: builds gitea → novox; no bus step; every machine composes with the change as it di…
mesh/repo-check head d4b8b2470159 merged into trunk aee97bf8ad42 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
A feature's diff is megabytes, too large to type into a call. gitea_commit now also takes patch_file, the
random hex name of a file in the module's spool directory /var/lib/mesh/forge-patches (mode 0733, owned by
the machine's account), with patch_sha256. The tool reads only that directory, never through a symlink or a
hard link, refuses a file whose SHA-256 differs, applies it as it applies patch, and removes it.
Co-Authored-By: Claude <noreply@anthropic.com>
mesh/merge-gate head f75e1a91d957 merged into trunk a95b3fa053e0 — pass: builds mesh-knowledge → novox; no bus step; every machine composes with the change…
mesh/repo-check head f75e1a91d957 merged into trunk a95b3fa053e0 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/merge-gate head c2cd2e1eb15e merged into trunk 789b0a9474c8 — pass: builds mesh-mail, messenger, new: modules/mesh-mail-notifier → novox; no bus step;…
mesh/repo-check head c2cd2e1eb15e merged into trunk 789b0a9474c8 — pass: THE CHANGE ALTERS ITS OWN CHECK (merge-check.sh): main's version judged it; the ch…
A session that only reset after a quiet minute kept counting, and so folding, a steady
trickle of mail forever even with no burst (one mail every 50s). N is now the account's
qualifying mail arrived in the trailing 60 seconds, counted fresh on every arrival; a fold
holds only its own overflow and the notice names just that count, not the whole history.
A dropped tell was kept in the same bounded send history (KeptSends)
that the conditions' and notify's cap counts sends from. A flood of
drops could push real, still-recent sends out of that bounded history,
undercounting sentLastHour and letting the conditions' cap stop
holding. A dropped tell is still said once an hour in the log and
named in Tell's own "dropped" answer; it is simply never recorded into
that shared history any more.
A buffer was replaced, discarding whatever it still held unsent, once
Window had passed since its first mail — losing a failed send waiting
for its retry, or a folded mail waiting for its minute's end, if a new
mail arrived meanwhile. The window now slides on the gap since the
last mail, not on time elapsed since the first: a session's count
never resets, and its buffer is never replaced, while anything in it
is still unsent.
The fold notice now counts every mail of the account in that minute,
not only the overflow, and says "mail" for one and "mails" for more.
ADR 0227 rule 8 / ADR 0240 rule 2 refuse a module with no health check of
another kind beside a tool check. mesh-mail-notifier has one process
(notify); wiring mail_notifier_health as its manifest health, as the review
note's wording suggested (mirroring mesh-mail's watch/mail_health), trips
that rule because mesh-mail also has keep (kind unit) beside watch.
mail_notifier_health stays built, tested and callable; notify's manifest
health stays kind unit until the lead decides how a one-process module
gets this automatically.
Flush now sorts accounts instead of walking a map, so notice order is
stable. A failed node-notifier.send keeps its mail pending and retries
at most once a minute, without holding up other accounts' notices, and
the failure is kept in the Store for a new health tool
(mail_notifier_health) that is unhealthy past ten minutes of failure.
The burst rule now follows the lead's reading of ADR 0378 S9: within an
account's sliding minute, the first three qualifying mails are shown
one by one and the rest are folded into one notice at the window's
end; there is no separate quiet-gap timer. The dedupe key is the
account and message id, falling back to folder/uidvalidity/uid only
when no message id is given, so a UIDVALIDITY reset cannot suppress a
new mail, and a redelivered event already buffered is never recounted.
status now lists every granted account, not only those with a notice,
and caches mail-store.accounts for a minute.
Review round 1 of mesh-catalog found sentLastHour counting tells
against CapPerHour (ADR 0380 §8 says tells keep their own cap, never
shared), and the drop log firing on every drop instead of once an
hour per channel as the comment already said.
ADR 0380 §8: the operator channel's notify is shaped for conditions (a key that
stays open, escalation, a plain rule that rewrites a subject), so a module has
no way to say a plain notice such as new mail. tell sends a headline and one or
two lines at once to every channel, bypassing routing, the burst window and the
walk hold; it offers Details like any message, is kept in the history, but its
record is marked cleared at once so it never stays open. Its cap (10 a channel
an hour) is its own, apart from notify's and the conditions'.
ADR 0378 §9 as amended by ADR 0379: a module of its own, built from mesh-mail's
event shape, runs as its own account to hear mesh-mail's restricted MAIL stream
(ADR 0376 §2), shows each new inbox mail through node-notifier.send at most
once, folds more than three from one account into one notice, and emits its
own shown event.
Declares mail-store and mail-webclient (held once for the mesh) and mail-notifier
(held once per machine, needing node-notifier) per ADR 0378 as amended by ADR 0379.
Claims mail-store and serves its four verbs through the same handlers as the
existing mail_* tools. Moves the event shape (EventKey, Changes) into its own
file, mailevent.go, so mesh-mail-notifier (issue 800 part B) can carry a
byte-identical copy. Adds checks/mailstore, refusing any manifest but mesh-mail's
own that claims mail-store (ADR 0378 §3), wired into merge-check.sh.
mesh/repo-check head eb97a3d9a2d7 merged into trunk 0914a33fba5a — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
A send whose ask never reached the router (the publish failed, or it was otherwise never recorded) stayed kept and
undecided forever, so mail_send_status kept saying "waiting" and the minute sweep kept reading it. Settle it as
ended once its asked time plus SendAskLasts has passed, with the words that say why. Keep an index of undecided
send ids beside the sends (written when kept, removed once decided) so the sweep's work grows with undecided
sends only, not with every send ever kept.
No tool named which units or processes used CPU outside containers during
the novox overload (issue 739). cpu_top reads each cgroup's cpu.stat (no
root needed), ranks units or containers busiest first, and says the
machine's CPU in use and the busy time no unit accounts for.
A delivery keyed by commit stayed stopped forever once an earlier, closed
duplicate pull request with the same head had stopped it, even though a
later pull request went on to merge that very commit: the merge only set
merged_as on the stopped record, and the walk that carried it waited for
this owner's word with no delivery left to give it. PullUpdated and
PullMerged now start such a delivery over, under the same id, for the
pull request that announces or merges it next, naming both pull requests
in the transition's why. Any other stop (a person's, a group's, the same
pull request's own close) is untouched.
The desk's prompt named only the module and the secret's key, so the operator
asking for one of mesh-mail's mail-credential-* accounts could not tell which
one. node-launcher.secret (modules/rofi) now shows the controller's optional
what as its own line under the module and secret names, escaped as text so
the catalogue's words never reach the prompt as markup; omitted as before
when none is declared. mesh-mail's mail-credential-* family now declares that
line with the long-form own-secrets shape the controller's manifest parser
reads (novox/hq issue 789 part b; part a landed in mesh-controller).
mesh/merge-gate head b5417c28d026 merged into trunk 5f6a41146c3a — pass: builds mesh-mail, sent nowhere; no bus step; every machine composes with the chang…
mesh/repo-check head b5417c28d026 merged into trunk 5f6a41146c3a — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/delivery covered after its stop: a later walk that contains it was delivered: plan-1792054175111921309 (novox/mesh-catalog at 0914a33f into main, do…
mesh/merge-gate head 2e32f47e4596 merged into trunk 9492278d934b — pass: builds gitea → novox; no bus step; every machine composes with the change as it di…
mesh/repo-check head 2e32f47e4596 merged into trunk 9492278d934b — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/delivery covered after its failure: a later walk that contains it was delivered: plan-1792097717281845096 (novox/mesh-catalog at 7443938f into main,…
The forge answers a repository's protection rules to its administrator
alone, and the tools' account no longer is one (ADR 0365), so every
delivery building a module failed at the controller's build-source check
with a 403. gitea_branch_protection_get now reads the rules with the
administrator account (ADR 0375); it only reads, and the protection set
and every write stay refused to the tools' account. Issue 772.
Co-Authored-By: Claude <noreply@anthropic.com>
mesh/merge-gate head 373f33975077 merged into trunk 7913b56d66f7 — pass: builds mesh-delivery → novox; no bus step; every machine composes with the change …
mesh/repo-check head 373f33975077 merged into trunk 7913b56d66f7 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
A delivery its group stopped for a member delivered since waits for a later
walk of its repository that contains its merge (issue 747). When that walk
carried no living delivery, nothing gave it its word and it waited forever.
The delivery now takes its turn on the walk that answers its merge, in its
group's order. A walk owed such a word stays under the orphans' timer, so
past the grace without it `stalled` lists it with why (held, or the
controller's refusal), and a walk no living delivery holds names the ended
deliveries it carries.
mesh/merge-gate head 48f0662fdea2 merged into trunk 5ce2d07c7f47 — pass: builds gitea → novox; no bus step; every machine composes with the change as it di…
mesh/repo-check head 48f0662fdea2 merged into trunk 5ce2d07c7f47 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
The tools' token reached everything the administrator may do, so their boundary rested on their own guards.
The gitea module now makes a plain account, mesh-tools, with the admin login, gives it write on code, issues
and pull requests of every novox repository through one team, and mints the tools' token for it; the admin's
old token goes only after the new one is kept. The Go forge bundle sends that token, never the admin login.
The forge shows protection rules only to a repository administrator, so every check reads a branch's
protection from the branch. mesh-admin stays a ready author, so marks made before the move keep counting.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
mesh/merge-gate head ab1811d00357 merged into trunk c79337579d6b — pass: builds new: modules/kernel, sent nowhere; no bus step; every machine composes with…
mesh/repo-check head ab1811d00357 merged into trunk c79337579d6b — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
A restart ran restore then apply, so every change passed back through the
found mode and rotated the journal again. Reload applies the new value
directly; restore also drops its state; no rotation while the machine boots;
the unit is sandboxed.
With THP 'always' journald's mmapped journal files sit in 2 MiB folios, so each small
journal write dirties and writes back 2 MiB; on the control node that saturated the
root mirror holding the bus store and the store, and the controller lost its lease.
The module sets the mode from a per-machine setting (default always, the kernel's own,
so assigning it changes nothing until a machine sets madvise), verifies it took, rotates
the journal once so new files are mapped under the new mode, gives back what it found
when stopped, and serves kernel_huge_pages to measure file_thp per unit.
mesh/merge-gate head 2833a049aa0f merged into trunk c51357f8afd4 — pass: builds gitea → novox; no bus step; every machine composes with the change as it di…
mesh/repo-check head 2833a049aa0f merged into trunk c51357f8afd4 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
POST /admin/hooks on gitea 1.27.3 makes a default webhook unless
config.is_system_webhook is "true": a default webhook fires for no existing
repository, and the system listing does not name it, so every start made
another and no event reached the receiver. Its edit never changes a secret
or a kind, so the new system webhook is made first and read back, then every
other hook at the receiver is deleted: system, default, and each
repository's copy of a default one, which would send its events twice.
gitea_events says where the receiver listens, the hook as ensured and what
it replaced, the verified and refused deliveries and the polls' pace. When
nothing verified for ten minutes, since the start or the last delivery,
while the poll found two heads or merges no event said, it is logged loudly
once per silent spell and answered unhealthy; so are deliveries refused
since the last verified one.
mesh/merge-gate head 6f7f07635e02 merged into trunk 0f76b6e232a5 — pass: builds mesh-delivery → novox; no bus step; every machine composes with the change …
mesh/repo-check head 6f7f07635e02 merged into trunk 0f76b6e232a5 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
A group member whose walk failed in an outage stayed failed after a later walk
of its repository, containing its merge, was done; its group then stopped every
later walk of the other repository, and the catalogue could deliver nothing.
A failed or stopped delivery now reads covered when a done walk of its
repository, made after its merge, whose commit contains it (the forge's word,
as issue 362's containment), built every module it moves. A group counts it as
delivered, and a member the group stopped for it no longer holds a walk back:
the group stops a walk only for a member whose change is still undelivered.
A stop covered under issue 480 now likewise makes its group read reached, not
stopped.
While the forge is asked, the walk is held back rather than stopped, for at
most ten minutes from the first ask: a forge that keeps erroring or cannot say
then lets the walk be asked to leave the stopped merge out, as before. The
hold is said by waits, stalled and stalled walks. The pass reads kept walks
each tick, so today's stuck deliveries clear after a restart.
mesh/repo-check head 3e16c055be8e merged into trunk 0361c44a03a0 — pass: THE CHANGE ALTERS ITS OWN CHECK (merge-check-typescript.sh): main's version judged…
Outside the image, assemble the same node_modules in the check's own
workspace from mesh-tools' lock on the forge, so a module's TypeScript is
checked locally before a push.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
mesh/merge-gate head d1debdcce816 merged into trunk 0e7d9e74d22a — pass: builds gitea → novox; no bus step; every machine composes with the change as it di…
mesh/repo-check head d1debdcce816 merged into trunk 0e7d9e74d22a — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
The runtime starts the watcher and the tools as two processes over one kept
token. Both minted at once on 2026-10-14 and left two tokens named
mesh-tools; gitea refuses a delete by a name held twice, so every mint
failed and every forge tool with it. A replace now deletes each token by
the name by its id, from the account's whole list, and a mint holds a lock
file beside the kept token and reuses what another process minted.
The review of #300 found a batch of 25 whole messages could hold a gigabyte,
mail_accounts told ungranted callers folder names, counts and server errors,
an unreadable grants setting was refused without its reason, and the watch
read the password only to learn it was there.
mesh/delivery covered after its stop: a later walk that contains it was delivered: plan-1792011830935225963 (novox/mesh-catalog at 0361c44a into main, do…
The gitea module keeps one system webhook for every event family, made or
corrected at start with the admin account (POST /admin/hooks, forge 1.27.3)
and a secret it mints and keeps. Its receiver, at the docker bridge's
gateway, refuses an unsigned request before reading it, a body over 5 MB,
and a signature that does not verify; it answers a verified delivery at once,
drops one already heard by its delivery id, and publishes each as
forge.<kind>.<action> on the module's events (its own stream, ADR 0363,
follows issue 725). A pull request opened, pushed to or merged is announced
at once. The polls look every five minutes only while deliveries arrive, and
every 30 s again, saying so loudly, when one finds what no event said or ten
minutes pass without a delivery. The announced records keep both paths from
announcing twice.
mesh/merge-gate head 2130f9961291 merged into trunk c71c513027e5 — pass: builds new: modules/mesh-mail, sent nowhere; no bus step; every machine composes w…
mesh/repo-check head 2130f9961291 merged into trunk c71c513027e5 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
ADR 0359 asked for a stream of its own and a bus grant from the module's
grants; ADR 0363 built both. The events leave the shared EVENTS stream, and
one family pattern per change replaces the literal per-account emits, so a
new account changes no manifest.
mesh/merge-gate head baa5ef802e82 merged into trunk 7e3e62062bef — pass: builds gitea, mesh-delivery → novox; no bus step; every machine composes with the …
mesh/repo-check head baa5ef802e82 merged into trunk 7e3e62062bef — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
gitea now declares a setting, so every mergeable file of it must name its
keys; the runtime config reads no setting, and ready-authors reaches its
own file.
mesh/merge-gate head 5b20e419ec1f merged into trunk 7e3e62062bef — fail: builds gitea, mesh-delivery → novox; no bus step; a manifest the change touches fa…
mesh/repo-check head 5b20e419ec1f merged into trunk 7e3e62062bef — fail: its merge-check.sh failed: files asking for a setting without saying whether it is…
Review of #303: anyone who can comment could turn a stale label into a mark,
and a long thread hid the newest comment. The forge now reads the pull
request's whole timeline and takes a "ready at" comment only from the
accounts in the trusted setting ready-authors (default the mesh's own forge
account and the operator's), written after the label was last added and after
the last push; any other author is ignored and said in the log.
The operator wants a reviewed pull request merged without a session calling
the merge tool again after every recheck. The reviewer marks it ready (label
mesh/ready and a comment "ready at <full head sha>"); the forge's open-pulls
read carries the mark, and mesh-delivery merges through gitea_merge_pull_request
at that head, in a group's order once its composed check passed. A push after
the mark clears it; a failed check is said on the pull request and as a
stalled line.
mesh/merge-gate head d94764f00168 merged into trunk fadf5025cc5a — pass: builds build-agent, route-proxy → ace, g14, novox, shanks; no bus step; every mach…
mesh/repo-check head d94764f00168 merged into trunk fadf5025cc5a — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/delivery covered after its stop: a later walk that contains it was delivered: plan-1792148283013560219 (novox/mesh-catalog at 3db85a0d into main, do…
Go's build ID hashes every input package, so a change mesh-builder never reaches
still changed its bytes; without it, a rebuild of the same program is the same
file, and the check can say the module is unchanged.
ADR 0359 decided one module holds the operator's mail accounts. This is its
first stage: an imap adapter that keeps IDLE on the inbox and polls the other
folders read-only, indexes each message in its own database, keeps each
attachment once by its SHA-256 in its bucket, and says every added, changed or
removed message as a metadata-only event. Bodies and attachments are served
only by its tools, to callers the trusted setting grants names. Its health
waits for the account's own secret until it is given at the desk (issue 710).
mesh/merge-gate head dbe2949f8456 merged into trunk 482232e8c51c — pass: builds mesh-catalog → novox; no bus step; every machine composes with the change a…
mesh/repo-check head dbe2949f8456 merged into trunk 482232e8c51c — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
A branch built through the build verb is a real build, and placing it would make it the catalogue's
current version though the controller never registers it. Say which of the controller's rules are
mirrored and which need its records.
mesh/merge-gate head b7d9590fdc95 merged into trunk e1c8965ddbf1 — pass: builds messenger, telegram → novox; no bus step; every machine composes with the c…
mesh/repo-check head b7d9590fdc95 merged into trunk e1c8965ddbf1 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
Issue 372's fix with issue 639's one line edited an answered ask down to
'You chose X at HH:MM · headline', so the operator could not read back what
they approved. The headline stays the title, the summary, points, link and
Details stay the body, and how it ended and who acts on it become the note.
Telegram cuts the body from its end past 4096 characters, never the title or
the note.
mesh/merge-gate head 7ee2869d60eb merged into trunk e1c8965ddbf1 — pass: builds mesh-catalog → novox; no bus step; every machine composes with the change a…
mesh/repo-check head 7ee2869d60eb merged into trunk e1c8965ddbf1 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/merge-gate head 0b74c04aa89f merged into trunk e1c8965ddbf1 — pass: builds mesh-catalog → novox; no bus step; every machine composes with the change a…
Since the build role moved to node-build-agent (ADR 0190), the catalogue consumed only the retired
mesh-build-machine.built, so builds reached it only through the replay it asks for at its start: mailu
e1c8965d, built two minutes after the last replay, was never registered. Consume node-build-agent.built,
and pass over the seat's merge checks, tries, dry runs and failures, which the old seat never said.
mesh/merge-gate head 47ef2644aca1 merged into trunk e22326240833 — pass: builds build-agent, gitea → ace, g14, novox, shanks; no bus step; every machine co…
mesh/repo-check head 47ef2644aca1 merged into trunk e22326240833 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
Every merge check's judge is built where no pull request can reach (hq ADR 0358 §6,
issue 704): the build agent needs a directory beside its workspace, mounted at the
same path, that only judge builds mount. ADR 0358 §5 already holds in the forge's
holder; this test keeps it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
mesh/merge-gate head 8e821ef44992 merged into trunk cb7bf94a77ea — pass: builds mailu, sent nowhere; no bus step; 1 wait(s) for a person; every machine com…
mesh/repo-check head 8e821ef44992 merged into trunk cb7bf94a77ea — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/merge-gate head 07b9bd0ed44a merged into trunk 02427a60ef78 — pass: builds mailu, sent nowhere; no bus step; 1 wait(s) for a person; every machine com…
mesh/repo-check head 07b9bd0ed44a merged into trunk 02427a60ef78 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
Port 25 takes no login, so a client that tries is guessing passwords;
mailu's jail only matched Dovecot's failures and never banned them
(issue 708). Relay attempts stay unbanned: postfix already refuses them.
mesh/merge-gate head 8661bcfa2275 merged into trunk 50861c74f8b4 — pass: builds mailu, sent nowhere; no bus step; 1 wait(s) for a person; every machine com…
mesh/repo-check head 8661bcfa2275 merged into trunk 50861c74f8b4 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/merge-gate head 60495376551d merged into trunk 50861c74f8b4 — fail: builds mailu, sent nowhere; no bus step; 1 wait(s) for a person; a manifest the ch…
mesh/repo-check head 60495376551d merged into trunk 50861c74f8b4 — fail: its merge-check.sh failed: files asking for a setting without saying whether it is…
A message to a far host that is down stays deferred for up to five days,
and mailu's unhealthy now pages the operator, so one dead recipient would
page for days. Overdue deferred mail is unhealthy only when a lookup failed
here (a name-service failure, as in issue 687) or when it reaches three or
more domains; one or two far domains are said in the answer, not alarmed.
mesh/merge-gate head bf3e801f6700 merged into trunk 50861c74f8b4 — fail: builds mailu, sent nowhere; no bus step; 1 wait(s) for a person; a manifest the ch…
mesh/repo-check head bf3e801f6700 merged into trunk 50861c74f8b4 — fail: its merge-check.sh failed: files asking for a setting without saying whether it is…
On 2026-10-14 every mailu container read healthy while postfix could not
resolve a name: inbound mail was refused with 450 and outbound mail sat in
the deferred queue for days (issue 690). The smtp container is now judged
by a tool check, mailu_health, asked inside it: postfix runs (the image's
own check, kept), no message waits deferred longer than 30 minutes, and a
public mail host resolves through the resolver postfix is given.
mesh/repo-check head 2bb5d344f6f1 merged into trunk af5cf53a3450 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/delivery covered: a later merge that contains it was delivered: novox/mesh-catalog@054dc9618ce3 (merged as b15846dd into main, walk plan-17919811644…
Any route umami does not serve answers 404 too, as an HTML page. The
share route answers JSON with code not-found only after the database
was asked; with the database down it answers 500.
mesh/merge-gate head 0fcc560fecd6 merged into trunk e846c289c878 — pass: builds lab → shanks; no bus step; every machine composes with the change as it did…
mesh/repo-check head 0fcc560fecd6 merged into trunk e846c289c878 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
A run and a rebuild each asked the package registry between checking for
the other and writing their status, so both could start and a bed could
raise machines from an image being replaced. The registry is asked first
now. A hard-killed rebuild's credential is cleaned like a run's; a job that
never recorded its process is lost rather than blocking every run; and
lab_check reports a missing or unreadable base image.
mesh/merge-gate head acf8a76646de merged into trunk ebf7f3773dc7 — pass: builds lab → shanks; no bus step; every machine composes with the change as it did…
mesh/repo-check head acf8a76646de merged into trunk ebf7f3773dc7 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
A test's own output, printed before the suite reports, could carry a line
shaped like the suite's NOT PASSED or skipped line and stand for its verdict.
mesh/delivery-group group fix/145-a-consumer-is-held-under-its-provider checking: composed check out of date: novox/mesh-catalog@b62bb158398d was judged agains…
gitea, nextcloud, umami and n8n gain an http check that reaches their
database through their own endpoint; mesh-knowledge's watch already
asks the database and now names it. While postgres is unhealthy, their
findings are held under its condition instead of raising one each.
mesh/merge-gate head ff9d3ee6a06c merged into trunk ebf7f3773dc7 — pass: builds lab → shanks; no bus step; every machine composes with the change as it did…
mesh/repo-check head ff9d3ee6a06c merged into trunk ebf7f3773dc7 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
A bed machine carries the package database its base image was built with,
and the mirrors stop serving it within weeks: genesis-keeps-its-bus failed
on a 404 from an image 44 days old, with nothing saying its age and no mesh
tool to rebuild it. lab_check and lab_run now say the age, lab_run refuses
an image older than 14 days and names the remedy, and lab_base reads the
image and rebuilds it with mesh-lab's base build when asked.
mesh/merge-gate head c8255eb15d12 merged into trunk ebf7f3773dc7 — pass: builds lab → shanks; no bus step; every machine composes with the change as it did…
mesh/repo-check head c8255eb15d12 merged into trunk ebf7f3773dc7 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
lab_status took the suite's exit as the verdict, and the runner exits 0 when
every test skipped: a genesis bed that skipped its only test as NOT REACHED was
recorded as passed. A run that ended now has its outcome read from its log and
kept in its status - totals, each skip with its reason, and why it did not pass
- and one whose exit said 0 while no test passed reads as failed. Read from the
lab's own lines when it prints them, from the runner's summary for a lab of
before.
mesh/merge-gate head 4ded07bc635f merged into trunk b15846dd6e78 — pass: builds mailu, sent nowhere; no bus step; 1 wait(s) for a person; every machine com…
mesh/repo-check head 4ded07bc635f merged into trunk b15846dd6e78 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
The image's own check asks nginx's /health and that dovecot runs, so the
front stayed healthy while mail could not flow (issue 688). It now also
needs port 25 to greet over SMTP and the served certificate to hold for
14 more days; certbot renews at 30, so a renewal stuck for two weeks is
said long before the certificate expires.
mesh/merge-gate head 054dc9618ce3 merged into trunk a44d490d974c — pass: builds mailu, sent nowhere; no bus step; 1 wait(s) for a person; every machine com…
mesh/repo-check head 054dc9618ce3 merged into trunk a44d490d974c — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
The root zone signs with KSK-2024 since 2026-10-11. The pinned resolver
image (2024.06.59) trusted only KSK-2017 and kept its anchor inside the
container, so every lookup failed with SERVFAIL and mail was refused or
stuck (issue 687). 2024.06.63 refreshes the anchor with unbound-anchor at
every start and fails its health check on SERVFAIL, which the old
'dig @127.0.0.1' never did (issue 688).
mesh/merge-gate head c876ad4a8303 merged into trunk 23a2e1e20467 — pass: builds build-agent → ace, g14, novox, shanks; no bus step; every machine composes …
mesh/repo-check head c876ad4a8303 merged into trunk 23a2e1e20467 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/merge-gate head 12a5e7be7321 merged into trunk 23a2e1e20467 — fail: builds build-agent → ace, g14, novox, shanks; no bus step; head 12a5e7be7321 does …
mesh/repo-check head 12a5e7be7321 merged into trunk 23a2e1e20467 — fail: head 12a5e7be7321 does not merge cleanly into trunk 23a2e1e20467: in conflict modu…
mesh/merge-gate head 66d22dd53eff merged into trunk 7ed119995f69 — pass: builds mesh-delivery → novox; no bus step; every machine composes with the change …
mesh/repo-check head 66d22dd53eff merged into trunk 7ed119995f69 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
A check's builds for its merge have no taker once the pull request closes
unmerged, and the delivery should say whether a module was reused from the
check or built, and why (novox/hq ADR 0350 §4, §5).
mesh/merge-gate head 1faee6e27976 merged into trunk faebc3569acf — pass: builds gitea, lab → novox, shanks; no bus step; every machine composes with the ch…
mesh/repo-check head 1faee6e27976 merged into trunk faebc3569acf — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
The forge writes every tarball at its public name and the lab reaches it at
a private address, so npm 12 refused every locked @novox package. The public
name now comes from the mesh's own composition of the forge's route, through
the lab's binding, and the lab compares scheme, host and port.
mesh/repo-check head f09daabec3a8 merged into trunk 9c166398f204 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
The router's fallback said a refused note line as "(a line carrying … is withheld here)", so the operator
could not tell which setting held a value they were not shown. A change's line now keeps its key; any other
line, or one whose key is refused too, is said withheld whole. The SDK pin moves to mesh-sdk 5933eb7, where
no two asks digest the same.
mesh/repo-check head bb5a342863ee merged into trunk 9c166398f204 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
The review of mesh-controller #304: a proposal of more than three keys showed only the names of the rest,
and its Details could carry no value, so Approve bound values the operator never saw. The router now
shows an ask's WholeDetails in the note in place of its Details on a kind that proves who answers, held
to the secret rule alone, and refuses an ask whose whole details carry a secret's shape. Telegram keeps
the note whole when a message is too long, cutting the body first and saying so, and escapes the note
as every other part. SDK at mesh-sdk f474544.
The operator would not read the Telegram ask for ADR 0343: thirty plain lines, every clause on one, no bold,
no summary. A settings proposal was the same wall of text, its values masked so the approval was blind.
- messenger: an ask's message is the asker's headline and explanation alone; who asked and when, what each
answer does, the expiry, where else it is answered and the asker's Details travel as the message's note
(asks.Message.Note). Answered, the message is one line, "You chose Approve at 16:06 on Telegram ·
<headline>". An explanation that does not read at a glance (asks.Glance) is shown and logged. On a kind
that proves who answers, an ask's values are shown whole (asks.Ask.Whole), held to the secret rule alone
(from the earlier work on issue 383, #178).
- telegram: every message in HTML, never reworded: the title bold, a key line monospace, the note apart in
a collapsed blockquote; markup escaped. A tap is compared by its words, whitespace aside, and a message
shown before is still compared by the digest it was kept under.
- desk-channel: the note follows the body.
- mesh-knowledge: a record that needs the operator carries a brief (headline, summary, at most three
points), the 30th per-record rule; its ask shows the brief and never the clauses. An ask made before is
composed again as it was asked (form), so its warrant still holds.
- claude-code: an instructions proposal's explanation is one summary line (the proposer's, or the
module's), its text in the Details; a question's context is at most 300 characters and its recommendation
one point. Older proposals and questions keep their form.
The SDK is pinned at mesh-sdk 9588871 (fix/639-an-ask-reads-at-a-glance), which adds Message.Note and
asks.Glance.
mesh/merge-gate head f1f866050eee merged into trunk e209d46faf3d — pass: builds betterbird, sent nowhere; no bus step; 2 wait(s) for a person; every machin…
mesh/repo-check head f1f866050eee merged into trunk e209d46faf3d — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/merge-gate head e46eeb829853 merged into trunk 4d15a155e634 — pass: builds claude-code → ace, g14, novox, shanks; no bus step; every machine composes …
mesh/repo-check head e46eeb829853 merged into trunk 4d15a155e634 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
The runtime hands a watch its current values one at a time and gives them ten
seconds. The handler read the state, waited for a running look and wrote files
for each value, so with about 450 memory values the watch never started on g14
or shanks, and a file written on g14 never reached shanks. The handler now
notes each key's newest revision and returns; a worker applies what was noted
at the state's current version.
mesh/merge-gate head 73404a3c3f21 merged into trunk 5f491e0c097d — pass: builds betterbird, sent nowhere; no bus step; 2 wait(s) for a person; every machin…
mesh/repo-check head 73404a3c3f21 merged into trunk 5f491e0c097d — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
With calendar.alarms.showmissed on, Betterbird's alarm service fires every
reminder of the past month that carries no later acknowledgement at each
load of a calendar. Owl never sends a dismissal to Exchange and keeps it only
in Betterbird's local cache, so on g14 September's Exchange meetings came
back about every ten minutes, Dismiss or not. The policy now locks the pass
off whatever the setting or profile says, and betterbird_check says when the
running Betterbird still has it on beside an Owl calendar.
mesh/merge-gate head 8ad121d5c369 merged into trunk 305c2ba24733 — pass: builds lab → shanks; no bus step; every machine composes with the change as it did…
mesh/repo-check head 8ad121d5c369 merged into trunk 305c2ba24733 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
What the registry answers may come over plain HTTP, so it must not choose
where packages come from, nor reach the run's npm configuration as a raw
string.
mesh/merge-gate head 305601d25ef4 merged into trunk a84d0094a0a3 — pass: builds claude-code → ace, g14, novox, shanks; no bus step; every machine composes …
mesh/repo-check head 305601d25ef4 merged into trunk a84d0094a0a3 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
A value in the state could name a project such as ../../.. and be written outside
the memory folder; a file or folder that could not be read was carried as a
removal; text kept over a removal or merged in a conflict reached the state
unscreened; and a conflict whose file could not be written was merged into the
state again on every look. Names are judged before any file is touched, what
cannot be read is kept and listed, every state write is screened, a conflict is
written to the file before the state and warns once when it cannot be, a file
changed between reading and writing is not written over, non-UTF-8 text is not
carried, and a node that was away takes a removal it holds no change against
(ADR 0343 §9, as clarified).
mesh/merge-gate head 910bc58fe01d merged into trunk a84d0094a0a3 — pass: builds claude-code → ace, g14, novox, shanks; no bus step; every machine composes …
mesh/repo-check head 910bc58fe01d merged into trunk a84d0094a0a3 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
A merged file could exceed the state's 256 KiB value; the newer now stays in the
file and the older is written beside it, each carried. A node that was away takes
the newest version when its own file holds no uncarried change (tested).
mesh/merge-gate head 0096beeff999 merged into trunk a84d0094a0a3 — pass: builds claude-code → ace, g14, novox, shanks; no bus step; every machine composes …
mesh/repo-check head 0096beeff999 merged into trunk a84d0094a0a3 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
The warning now opens with what the operator is needed for and names no path,
agent or session (ADR 0253's plain rule); clause references follow ADR 0343 as
reviewed.
mesh/merge-gate head 6aa30eae66be merged into trunk a84d0094a0a3 — pass: builds claude-code → ace, g14, novox, shanks; no bus step; every machine composes …
mesh/repo-check head 6aa30eae66be merged into trunk a84d0094a0a3 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
A session started on another node began without what the sessions elsewhere had
learned, because Claude Code keeps project memory on one node. The module's own
process now carries the memory folders through its kept memory state: a change
is written elsewhere only over the version it replaced, a file changed on two
nodes keeps both (newer first) with a warning, and every replaced or removed
version is kept. No tool writes memory.
mesh/merge-gate head bb60de6d616a merged into trunk a84d0094a0a3 — pass: builds lab → shanks; no bus step; every machine composes with the change as it did…
mesh/repo-check head bb60de6d616a merged into trunk a84d0094a0a3 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
npm 12 installs a locked tarball only from its configured registry; the forge
writes every tarball URL, and so every lock file, at its public address, and
the lab named the registry by its forge setting's address. Every run stopped
at mesh-tools' install with EALLOWREMOTE.
mesh/merge-gate head dd39d4ace74b merged into trunk 0d723484a84a — pass: builds lab → shanks; no bus step; every machine composes with the change as it did…
mesh/repo-check head dd39d4ace74b merged into trunk 0d723484a84a — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
The lab's tools read their bound npm-package-registry credential, root's at 0600 on the machine. The
controller gives the runtime's account exactly the files a tools bundle names in its env (ADR 0192), as
gitea and umami are given theirs, so the bundle now names the credential and its binding there, and no
other credential changes hands.
The merge check runs as root, where no directory mode stops a removal: the test that makes an
unremovable tree now skips there before it makes anything, and a second test clears leftovers whoever
runs it.
mesh/merge-gate head 5bd40fa4061d merged into trunk 0d723484a84a — pass: builds lab → shanks; no bus step; every machine composes with the change as it did…
lab_check cloned into one fixed path, <work>/check, after an rm -rf whose result it ignored; a tree an
earlier check left that the lab's account could not remove made every check fail on "destination path
already exists". Each check now works in a fresh check-* directory it removes when done, and clears
what earlier checks left, naming any path it could not remove.
mesh/merge-gate head 8d8dcd5ebb86 merged into trunk cd239283207a — pass: builds gitea, lab → novox, shanks; no bus step; every machine composes with the ch…
mesh/repo-check head 8d8dcd5ebb86 merged into trunk cd239283207a — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
The lab's credential came from the package team, which also writes the @novox packages, and the
branches a run clones are not reviewed yet. The lab now contributes {"access": "read"} to its
npm-package-registry requirement, and gitea's provisioner puts such a consumer on a team of its own,
readers: code read on the org's repositories, package read, no write, and off the package team. The
build agent, which publishes, stays on packages.
A stop's SIGTERM to the run's process group killed the wrapper before it removed the credential file.
The run's script now removes it as soon as its clones are done and traps every way it ends, and a run's
start and lab_check remove what a crash or a reboot left.
mesh-tools-go is private, and the lab cloned over HTTP with no credential: a run failed at its checkout,
"could not read Username". The lab now requires the package registry, as the build agent does; the
forge's provisioner binds it an account and seals that account's password, and a run offers them to
the forge's own origin through git's credential store, from a file beside the run and removed when it
ends: never in a URL, the run's script, an argument or the log. lab_check reads every repository a run
clones with that same credential and names one it cannot read.
mesh/merge-gate head 1725ffc799bd merged into trunk 8922c890595b — pass: builds claude-code, claude-licence-manager → ace, g14, novox, shanks; no bus step;…
mesh/repo-check head 1725ffc799bd merged into trunk 8922c890595b — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
The bindings a login makes are now announced as licence.bound before the adoption's own event, so the test asks for each event rather than for adoption to come first.
mesh/merge-gate head a72eb3dbb28d merged into trunk 19ecad261cf2 — pass: builds claude-code, claude-licence-manager → ace, g14, novox, shanks; no bus step;…
mesh/repo-check head a72eb3dbb28d merged into trunk 19ecad261cf2 — fail: its merge-check.sh failed: --- FAIL: TestAManagerWithNoLicenceAdoptsTheNewestLogin…
A switch of g14 back to another licence on 2026-10-13 left no trace: the manager logged no line and sent no event for bind, switch or release, and its audit kept only the verb's name. Every change of what a consumer holds now goes to the journal, the audit and a licence.<what> event with the same fields: consumer, licence before and after, generation, why (the verb or the rule) and by (the caller the runtime proved, or the manager itself). The node is told why in the bindings state and says it when it applies. The audit and binding tables become readable through the store's reader. The SDK moves to the version that proves the caller (issue 365).
mesh/merge-gate head c0726ac2641b merged into trunk 19ecad261cf2 — pass: builds new: modules/dev-tools, sent nowhere; no bus step; every machine composes w…
mesh/repo-check head c0726ac2641b merged into trunk 19ecad261cf2 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
Every node is a development machine: an agent builds a change and runs its tests there before opening a pull request, and the lab tests it afterwards as QA. No module declared Go or npm, so no node could do the first part.
mesh/merge-gate head c6a9c52dc7f2 merged into trunk 45d01e77c26d — pass: builds lab → shanks; no bus step; every machine composes with the change as it did…
mesh/repo-check head c6a9c52dc7f2 merged into trunk 45d01e77c26d — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
lab_run ran npm ci at the root of mesh-sdk, mesh-tools and mesh-lab; mesh-tools keeps its package in
node-tools/ with no lock file at its root, so every run failed at its setup in nine seconds. A run now
installs (and, for the SDK and the base, builds) where a package-lock.json is: the root, or each
directory one level down; a package.json with no lock file fails the run naming it. It checks out
mesh-tools-go, which genesis builds since issue 616, and tells the beds where its checkouts are
(MESH_LAB_REPOS). lab_check asks the forge for every repository a run clones and applies the same install
rule to the ones it installs, so a run's setup failure is found before a run.
mesh/merge-gate head c78d0a01342c merged into trunk f35b06750452 — pass: builds fail2ban → ace, g14, novox, shanks; no bus step; every machine composes wit…
mesh/repo-check head c78d0a01342c merged into trunk f35b06750452 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
fail2ban reads the ignoreip entry file:/var/lib/... and prints it back as file://var/lib/...,
in get <jail> ignoreip and in its "Ignore ... by ..." log lines alike. The watch compared the
text, so on ace it said every jail lacked the list while each held it, failing fail2ban's health
and its walk; and it would never have counted a failure from the operator's own address. Compare
the path instead, with a test on fail2ban's real form.
mesh/merge-gate head 93d58f3740a0 merged into trunk 5cd608ba8898 — pass: builds fail2ban → ace, g14, novox, shanks; no bus step; every machine composes wit…
mesh/repo-check head 93d58f3740a0 merged into trunk 5cd608ba8898 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/delivery covered after its failure: a later walk that contains it was delivered: plan-1791883115921573192 (novox/mesh-catalog at 45d01e77 into main,…
An agent polling the forge from the operator's home got that home banned for a day and then four
weeks, and nothing in the house reached the control-node. Every jail spared only private ranges.
The mesh-wide setting own-addresses reaches the machine as a mergeable file whose own content is
the empty list, so the module composes while it is unset. A new watch process (root) writes its
entries into a list every jail's base ignoreip names as file:, which fail2ban reloads without a
restart; it checks every jail names that list and keeps each failure a jail ignored because of it
for a day. The new fail2ban_health tool judges the watch, so the controller raises a warning
naming node, jail, address and count. The ban verb refuses an address on the setting or on the
jail's ignore list before the daemon is asked.
mesh/merge-gate head 2f5e97030a5c merged into trunk bc3b59227f46 — pass: builds mesh-delivery → novox; no bus step; every machine composes with the change …
mesh/repo-check head 2f5e97030a5c merged into trunk bc3b59227f46 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
A walk that moved nothing because every module's recorded build source held none of the changed files now
says so per module (delivery.unchanged). Such a merge moves nothing (ADR 0270 §1): it is delivered, naming
the modules, not stopped 20 minutes later with the operator asked about a change already on every machine.
A module the walk neither built nor said unchanged still stops it (issue 474).
mesh/repo-check head eb4280637906 merged into trunk 53e51864969f — pass: THE CHANGE ALTERS ITS OWN CHECK (merge-check.sh): main's version judged it; the ch…
grafana's client logged in with a token setting ahead of the vault's admin password,
and its runtime config let a layer set token. The token setting is now ignored and
said, token is protected, and the vault's password comes first.
checks/secrets/known could gain a line hiding a newly unprotected key. merge-check.sh
now judges it against the trunk's copy at the merge base, and refuses an added line.
mesh/repo-check head fe2040c65ea2 merged into trunk 53e51864969f — pass: THE CHANGE ALTERS ITS OWN CHECK (merge-check.sh): main's version judged it; the ch…
A setting outranked the vault: baserow, grafana, letta and unifi merged a layer's
password over the ${secret:...} in their runtime config, and the gitea,
home-assistant, icecast, influxdb, mailu and nextcloud clients read the settings
file before the vault's file. A secret in a setting cannot be rotated and is read
back by the settings verb.
The four files now keep their password protected, so the controller refuses a
setting of it. The six clients ignore a secret in the settings file and say so
once, by key and never with the value; icecast reads its admin password from the
vault's file. checks/secrets refuses a mergeable file whose secret a layer could
replace; searxng's nested server.secret_key is listed as known until the
controller protects a nested key.
mesh/merge-gate head e0bc919d4945 merged into trunk f88e3452f8b8 — pass: builds build-agent → ace, g14, novox, shanks; no bus step; every machine composes …
mesh/repo-check head e0bc919d4945 merged into trunk f88e3452f8b8 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/delivery covered: a later merge that contains it was delivered: novox/mesh-catalog@2f5e97030a5c (merged as 5cd608ba into main, walk plan-17918799551…
build-agent's npm-package-registry becomes optional, so genesis's step 13 is no
longer refused with 'nothing provides npm-package-registry'. Until gitea is
assigned, the agent on genesis's node reads the genesis-package-registry
own-secret the installer seals at the packages step; it is issued outside the
mesh, so no other machine is given one. The binding and its credential are
composed only while bound (ADR 0330 §6), and no file names them otherwise.
mesh/merge-gate head c9a711c259ed merged into trunk 827b89dbf109 — pass: builds betterbird, sent nowhere; no bus step; 2 wait(s) for a person; every machin…
mesh/repo-check head c9a711c259ed merged into trunk 827b89dbf109 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
writeAccountLists checked only for a running Betterbird and wrote without a sync;
it now waits for the profile's lock too and writes atomically. readLayer read any
answer that was not JSON as an empty layer, and a proposal replaces the layer
whole, so a changed wording would have dropped accounts; it now refuses. The
README says the rollout order that leaves no node on Betterbird's defaults.
mesh/merge-gate head 1835f0efad0f merged into trunk 39e7a06e6293 — pass: builds betterbird, sent nowhere; no bus step; 2 wait(s) for a person; every machin…
mesh/repo-check head 1835f0efad0f merged into trunk 39e7a06e6293 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
The module owned only the accounts' server fields, so identities, signatures,
calendars, address books, filters and tags could differ between desktop nodes
unseen, and the operator's own preferences were shipped in the catalogue.
prefs becomes one entry per preference (a node's layer overrides one key) and the
defaults file keeps no value of the operator's; identities, calendars,
address-books, filters and tags become settings, adopted like accounts, filters
owned whole per account, and what Betterbird rewrites written only while it is
closed. New tools list each kind, derive the layers from every desktop node and
propose one at a time, and show drift; every answer passes a credential check.
mesh/merge-gate head 84911b94f3be merged into trunk 8f80cd471634 — pass: builds lab → shanks; no bus step; every machine composes with the change as it did…
mesh/repo-check head 84911b94f3be merged into trunk 8f80cd471634 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh-host now carries one genesis bundle, foundation-first-node.lock; the -nats one the
lab's runs named is deleted, so every run would have raised from a file that is gone.
mesh/delivery-group group feat/335-a-busy-container-is-replaced-when-idle rejected: its order contradicts itself: novox/mesh-catalog@190608193764, novox/mesh-c…
Every merge into mesh-controller replaced the agents and the queue started
their checks again elsewhere (issue 594). Its server now declares busy as
mesh-builder busy (novox/hq ADR 0335 §5).
mesh/merge-gate head 4193eacec459 merged into trunk 39e7a06e6293 — pass: builds grafana, home-assistant → ace; no bus step; every machine composes with the…
mesh/repo-check head 4193eacec459 merged into trunk 39e7a06e6293 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
grafana runs without influxdb and home-assistant without the Servarr apps or a broker, yet a mesh
without them could not assign either. Their requirements are now optional: bound when a provider
runs, absent otherwise, with nothing installed for them. grafana's data source and its credential are
delivered only while influxdb-api is bound, into directories it mounts, so an absent file never
becomes a directory docker made in its place.
mesh/merge-gate head c4eebce598cb merged into trunk 9c9a76d220f8 — pass: builds claude-code → ace, g14, novox, shanks; no bus step; every machine composes …
mesh/repo-check head c4eebce598cb merged into trunk 9c9a76d220f8 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
Freeing the slot two minutes after asking could let a backlogged router still show
the question after the session read it as ended, and let a third question in. The
seat's queue offers work again after an error and the router drops a cancel of an
ask it has not recorded, so a cancel cannot take it back; once expired, the router
refuses it if taken late. The slot test now uses a stale revision, so a plain
write in place of the compare-and-set fails it.
mesh/merge-gate head 14999f1d3ecb merged into trunk 9c9a76d220f8 — pass: builds claude-code → ace, g14, novox, shanks; no bus step; every machine composes …
mesh/repo-check head 14999f1d3ecb merged into trunk 9c9a76d220f8 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
A question was kept on the bus before the router judged its words, so a token in
it reached a bucket every node reads: every word, the options' too, is now held
to the secret shapes and invisible characters before anything is kept, and a
refused question keeps none of its words. The cap was check-then-create and
failed open: each open question now holds one of two slot keys taken by create
or compare-and-set, refusing when a slot cannot be read. A question the router
never recorded frees its slot after two minutes instead of at expiry. by is
required; wording matches ADR 0332.
mesh/merge-gate head e4e0ce761901 merged into trunk 9c9a76d220f8 — pass: builds claude-code → ace, g14, novox, shanks; no bus step; every machine composes …
mesh/repo-check head e4e0ce761901 merged into trunk 9c9a76d220f8 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
The refusal test failed on a loaded build agent: Propose returned on hearing the
refusal before HeardEnd had written the discard, so a reader right after saw
nothing decided. The question path keeps the same order.
mesh/merge-gate head 9446bec2f969 merged into trunk 9c9a76d220f8 — pass: builds claude-code → ace, g14, novox, shanks; no bus step; every machine composes …
A session that needs a decision could only ask in a terminal the operator may
not watch. claude_code_ask raises the question under this module's name at the
level approve, so only a channel that proves who answered carries it;
claude_code_answer reads the chosen option back by its key from the router's own
record, held to the question kept in the new questions state.
mesh/merge-gate head d09dbe33ac0f merged into trunk ac2d4dc46f4c — pass: builds build-agent, route-proxy → ace, g14, novox, shanks; no bus step; every mach…
mesh/repo-check head d09dbe33ac0f merged into trunk ac2d4dc46f4c — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
Neither server declared health, so each walk that moved them waited out the
60 s grace on every machine. The build agent is ready when 'mesh-builder ready'
says it took its build work; the route proxy when its http port accepts.
mesh/merge-gate head 005f802504fc merged into trunk eb5f33f2b475 — pass: builds mesh-delivery, mesh-issues → novox; no bus step; every machine composes wit…
mesh/repo-check head 005f802504fc merged into trunk eb5f33f2b475 — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
mesh/delivery covered after its failure: a later walk that contains it was delivered: plan-1791834103351744736 (novox/mesh-catalog at ac2d4dc4 into main,…
mesh/merge-gate head c4b4bac757c5 merged into trunk 0863ec2a059d — pass: builds betterbird, sent nowhere; no bus step; 1 wait(s) for a person; every machin…
mesh/repo-check head c4b4bac757c5 merged into trunk 0863ec2a059d — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
Betterbird keeps the active theme in its add-on database, and extensions.activeThemeID only mirrors it.
So clearing the preference changed nothing, and the drift report then hid that. The theme is now read
from extensions.json against default-theme@mozilla.org. Another theme is drift, and its remedy is the
add-on manager: config_apply never clears it and keep refuses it.
A credential's, an account's or a server's name in the defaults file is refused by code: it is never
answered or cleared, and the check names it. Apply also refuses while the profile's lock is there, and
syncs prefs.js before the rename. A saved default of a built-in preference is no longer drift.
A module that installs Betterbird but leaves every node to its own settings does not set Betterbird up.
The operator approved the configuration on 2026-10-12, and it now ships in three layers. The policy
locks what must never differ: telemetry, add-on updates, the vendor's notices and the start page. A
defaults file carries the laptop's setup, so a fresh node gets it untouched. The setting prefs stays a
node's exceptions.
A default loses to a value the profile saved. betterbird_config shows that drift, betterbird_check
reports it, and betterbird_config_apply clears it while Betterbird is quit. keep proposes a node layer
instead of writing user.js. betterbird_prefs now reads Betterbird's real defaults from omni.ja, and
withholds of mailnews.* only what names an account. The default-client ask is decided from its own
policy entry, not from the first thing the policy lacks.
mesh/merge-gate head 01d2ff7a46bc merged into trunk 2fe5601397ae — pass: builds gitea → novox; no bus step; every machine composes with the change as it di…
mesh/repo-check head 01d2ff7a46bc merged into trunk 2fe5601397ae — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
The merge tool deletes the branch after merging; the forge then reads head.ref as refs/pull/<n>/head and only
head.label names the branch. pull.merged carried the pull ref, so the controller found no group for the merge,
cut mesh-catalog #251's walk at once and walked it before mesh-controller #271's, against the group's order.
mesh/merge-gate head 961e57f7e403 merged into trunk 22071dabfc7b — pass: builds gitea → novox; no bus step; every machine composes with the change as it di…
mesh/repo-check head 961e57f7e403 merged into trunk 22071dabfc7b — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
A pass read only the 20 most recently updated open pull requests of a
repository, so a new head beyond them went unannounced and its merge check
never started (novox/hq issue 579). Pages stop at the floor; a sweep reads all.
mesh/merge-gate head 481e44cdaa29 merged into trunk 23ea0f431faf — pass: builds betterbird, sent nowhere; no bus step; 1 wait(s) for a person; every machin…
mesh/repo-check head 481e44cdaa29 merged into trunk 23ea0f431faf — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
Betterbird's own default asks at every start with a window to become the default for mail, news, feeds and
calendar files. The policy locks mail.shell.checkDefaultClient off, which wins over a profile's own value, and
betterbird_check says when a node would still ask.
mesh/merge-gate head 2f55adc57c60 merged into trunk 43a9155d5afe — pass: builds betterbird, keycloak, searxng → ace; no bus step; 3 wait(s) for a person; e…
mesh/repo-check head 2f55adc57c60 merged into trunk 43a9155d5afe — pass: its merge-check.sh passed, each part in its toolchain: merge-check.sh (go), merge-…
Every setting reached every mergeable file: searxng's endpoints landed in
its settings.yml, keycloak's issuer in its tools' config, betterbird's
prefs would land in its accounts file. Each file now names the keys it
reads, so the controller lays only those over it and calls the rest
stray. Merge only after the controller that knows takes-settings
(mesh-controller #272) has walked: an older one hands the field to the
node-engine, which refuses it.
The merge tool's pending bypassed the merge check's one writer from the tools'
bundle (issue 456); it is now served beside it. A verdict no build judges names
the announced base head, so the merge tool no longer refuses it for ever. A
late verdict keeps its account, a trunk is read again in each head's turn so a
fresh verdict is not put out of date, and a delivered member counts when a
rebased member's group is composed again (novox/hq ADR 0323).
A merge check's verdict now names the trunk commit it judged (novox/hq ADR
0323, issue 341), but the forge kept it green after another merge moved the
trunk, the merge tool merged on it, and mesh-delivery kept the delivery ready.
The forge's holder puts both statuses of every open pull request into a moved
trunk back to pending with the reason and runs nothing; the merge tool compares
the verdict's trunk commit with the trunk's head, asks the check once and never
merges by itself; mesh-delivery sends a ready delivery back to proposed, counts
only a verdict for the trunk's head, and composes a group again for a rebased
member instead of holding it.
mesh/delivery-group group fix/403-fix-time checking: composed check out of date: novox/mesh-controller@7ca5a45ecc5a was judged against trunk ?, the trunk is no…
The forge refuses an approval of one's own pull request, so the independent
reviews are comment reviews opening with "Verdict: approve at <sha>"; read
only APPROVED, the review phase was never measured. A fix whose start was
read back from the log was fixed before fix times were measured: it is kept
and shown, never warned on. A carried value at fixing after one at diagnosing
starts at its worker's active (the move), not its since (the diagnosis).
mesh/merge-gate pass: builds mesh-delivery, mesh-issues → novox; no bus step; every machine composes with the change as it did without (4 of 4 compose)
hq ADR 0313: the operator's five minutes start when an agent moves the issue
to fixing, not at the merge. mesh-delivery says on each transition, and in
checks, when the pull request opened, when it merged and the class of what it
moves. mesh-issues keeps the first move to fixing, the fix time with its four
phases once the last fix is delivered, serves fix-times, and says
issue-tracker.fix.<class>.over-budget on the operator's channel, cleared by the
next fix of that class within budget.
mesh/merge-gate pass: builds betterbird, sent nowhere; no bus step; 1 wait(s) for a person; every machine composes with the change as it did without (4 of …
The shared-server check refused an adopted account for its own alias, and
counted Owl's, local folders' and feeds' identities as users of the default
server, though they never send by SMTP. The controller now judges an
upstream pin's host like any name, so each pin declares its vendor.
The user.js answer carried declared addresses and hosts (ADR 0316 §16). An
adopted identity's outgoing server may be shared with an account the module
does not own, whose sending the declared values would change at every start:
refused with the remedy, since ADR 0317 §2 leaves the identity's server and
the lists as they are; a recorded adoption follows the identity's current
server. Owl installed by policy cannot be removed, so a newer Owl than the
pin is a fact whose remedy is moving the pin.
A module that cannot set Betterbird up is of little use (the operator,
2026-10-12). The module now declares the community package absent and
unpacks the pinned upstream release where that package had it, so the same
profile opens; places the enterprise policy that turns Betterbird's updater
off and force-installs Owl from the xpi the mesh pinned by its vendor's
digest (never in a repository); and takes the operator's accounts from the
trusted setting accounts, without any password or token: an account the
profile already holds is adopted under its own keys, a new one is created
under the module's mesh_ keys, an ambiguous one is refused with the remedy,
and the account lists are only written into. betterbird_accounts says the
profile's accounts by count and proposes the setting built from them for the
operator's Approve. Every new build waits for the operator's push. The
reminder test is visibly a test (MESH TEST band, event name) and its
description says it opens a window on the operator's screen.
A fresh profile has never checked for updates, so Betterbird checked at the
test's start, found a newer build published and opened its About window on the
operator's screen instead of leaving the reminder to show. The test's own
preferences now also come last in its user.js, so nothing declared or carried
overrides them.
No record names an image the mesh did not make, so one a definition pins under an
older name of the store read as unrecorded, the same word as an image nothing
uses. The tool now reads every definition through the catalogue and says
named-by-a-definition with the module. The store's older names are the mesh's
setting older-store-names, since a catalogue names no installation.
mesh/merge-gate pass: builds new: modules/mesh-reads, sent nowhere; no bus step; every machine composes with the change as it did without (4 of 4 compose)
Reading a file, searching one, listing a directory, the owner and mode of each
component of a path and looking up an account went through the login shell's
execute as the operator's account, which sees the operator's keys and the
agent's credentials. This module holds node-reads on every node and runs as
mesh-reads (no login, no group beside its own). It refuses the node's secret
places by name, judging the file actually opened rather than the name asked,
so a flipped link reaches nothing refused; withholds any line shaped like a
secret through mesh-sdk's outward rule and every line of a BEGIN/END block;
and bounds every answer. Replaces mesh-catalog #244 (reviews 55), whose first
commit carried a built binary.
delivery_status said only what is owed now; issue 326's rule needs the tries made in a window.
It now says the tries since this owner started, by how each ended, and the last try's time.
(novox/hq issue 407)
A changes link onto a drafted clause was refused, so records that really
change such a rule could only cite it (ADR 0294, 0301, 0320); and return
renumbered ADR 0281's clauses, pointing other records' links at other
clauses. The changed drafted clause keeps its number and stays unserved;
return refuses a record whose clauses another record changed or links by
number. Issue 494, ADR 0321.
A stop for a walk that never built its modules (issue 474) kept saying "error: stopped" on the forge and in
show after a later delivered merge carried its change to every machine. The stop stays final, as a covered
superseded member stays superseded (issue 442); its status, view, note and show now say covered and by which
merge, told once and from what is kept, so a restart or a stop covered before this tells it too. deliveries
dropped such a stop after a day while stalled still listed it; it now lists it as long as stalled does.
novox/hq issue 480, points 1 and 3.
A replaced ask's message was left open as if the newer ask took it over, but
the newer ask opens a message of its own, so the old one kept offering answers
nobody takes. It now ends on every channel with 'Replaced by a newer question',
and the minute's sweep ends any such message left from before. A record that
clears no longer keeps what a walk owed of it (the review of PR 238).
A warning held on purpose and lost with the walk's message at a restart was turned
into history and never told. Name the walk in the told line from its own item.
A resolved issue's worker proved the acceptance live, so a fix whose own walk failed
and that a later merge carried is not waited for. Only wontfix ended the warning, so
16 warnings stayed open on the operator's channel and the sweep raised new ones for
resolved issues.
A message lost with a restart before it was sent left what it carried marked told,
so it reached the operator only as its reminder hours later. A later walk took over
silenced, already said or asked messages that ADR 0308 section 3 never holds. A start
capped at the router's now was compared with the controller's moments.
The operator decided on 2026-10-12 that a rule of the record set that fails
is a fault of the records, which their authors repair, not the module
failing: as issue 508 decided for a proposal never reviewed, health says
only whether the module works, and a finding must not hold the module's
own rollouts. The findings go through the notice the knowledge base
already sends (operator-channel.notify, a warning), once per finding,
kept as told across restarts, and cleared when the rule holds again.
inZone wrote time.Local while the bus test's nats server read it from a
goroutine that outlives its test; the race detector the repository check
now runs failed TestAReviewOwedOutlivesARestartAndItsLastTryIsSaid.
The whole-set rule cited ran only after acceptance: the review approved a
decision with no home and no record linking it, the rule then failed, and
the module's health turned red for a fault its author could only repair
after the fact. hq held the rule at the record's own merge. The review now
judges it ahead, before the judgement: such a decision stays proposed with
a changes finding naming the rule. It is not refused at propose, because
no record can link a number only reserved.
Review 30 found guard.ts missing branches the forge protects: Go reads \d, \w,
\pL, a class opening with ] or ^, and [[:digit:]] otherwise than JavaScript.
A rule using any of them now covers every branch, so the guard refuses more
than the forge, never less. test/rule-cases.json holds the forge's answers,
read by the Go and TypeScript tests alike.
The forge's contents and diffpatch endpoints take a branch, never a commit to
build on, so the parent is read from the commit the forge made, and a branch
that moved under the commit is said. Blobs are read at the commit read first,
not at the branch name. Protection rules match as Gitea 1.27 matches them
([...], {a,b}, escapes, a plain name in any case), in the Go bundle and in
guard.ts alike. One commit carries at most 4 MiB per file or patch and 16 MiB
together; code search stops at its deadline and says so. The HTTP client the
two Go bundles copied is one package, internal/forgehttp.
The recheck's pending was written from the tools' bundle, its own process, past the
writer's per-head chain: the first run's verdict mark stayed set, so a rechecked head
never said where its new run builds, and an old run's verdict in flight or redelivered
could land over the pending. The status tool is now served from the watcher's bundle
and writes the merge check's pending through the writer, and a verdict whose ask
(checked.id) predates the check now pending is dropped, weighed after a restart against
the forge's own history of the gate (novox/hq issue 456).
mesh/merge-gate pass: builds nats, sent nowhere; merging this needs a person's bus upgrade; 1 wait(s) for a person; every machine composes with the change …
nats_user_can read the user list on the machine. A list written is not a list the bus enforces until
the server reloads it: on 2026-10-09 the controller asked between the write and the reload's effect.
The answer now carries in_force — allowed in the file, and the server's config_load_time after the
file's write — with why when it is not, and anything not known is not in force.
- No channel ready at a walk's end or bound: what it held is left unsent and tried
again every minute, as any other message, instead of lost with the walk.
- At a walk's end or bound, a message another open walk touches is held on by
that walk, even one started after it was held (hq ADR 0308 §2).
- A walk read back past its forget is released at its bound first, then forgotten.
- A walk's end is kept a day: a start heard after it (same started_at, or not
later than the end) opens nothing, also after the router restarts; a later
start under the same id still opens.
- A walk's plain end, when its words cannot be said, leaves what it held to be
said on its own instead of marking it told.
- The restart pass no longer skips a held message: one with an open ask is never
held (ADR 0308 §3), so its answers are made to work again.
An agent could not get a finished change onto the forge without someone
pushing by hand, so the branch / pull request / approval rule could not be
followed. A Go bundle serves gitea_commit, held to gitea_api's rule (never the
default branch, never a protected one), and the summarised reads agents
otherwise made through gitea_api as raw JSON.
mesh/repo-check pass: THE CHANGE ALTERS ITS OWN CHECK (merge-check.sh, merge-check-typescript.sh): main's version judged it; the change's judges the pull r…
A module whose dependencies could not be fetched was said NOT TESTED and the
repository check passed, so a merge on it would have landed untested code.
The build seat clones the SDK beside every check already: a forge requirement
the module cache lacks is now taken from that clone through a go.mod of its
own, and a module that still cannot be built fails the check by name and why.
A TypeScript module whose dependencies the toolchain lacks fails the same way.
A delivery's own restarts reached the operator as notifications while it ran, and every restart of the
tool runner popped the four silenced messages up again on the desk, twice. The router now reads the
controller's walk start and end on plan-moved, holds what an open walk touches until its end or bound,
tells the end as one message, and makes the restart pass once per restart from its own state, leaving
silenced, withheld, answered and held messages as they are.
The review showed the give-the-word rule skipped a failed or missing check
and group order, and let go moves nothing judged; to-be 47 has no such path.
Such a walk is now said, by stalled with walks, as held by nothing, why.
A merge whose plan sends nothing is delivered at once, yet the controller
can still open a walk for it; no live delivery then held the walk and it
waited for a person (the nzbgeek case). Give such a walk its word.
Name each held delivery in the status, and let stalled say, for every
waiting walk, what holds it, so the controller's plans line points at the
person's decision instead of a hand start.
An open's lost answer is judged by the issue's first entry; an entry that does not
fold faults its issue on start; a value written beside the log is never folded over;
one issue is kept in at most 252 KiB so its whole operation fits a log entry.
mesh/merge-gate fail: builds mesh-issues → novox; no bus step; a manifest the change touches fails the module check: modules/mesh-issues/module.json: this …
mesh/merge-gate fail: builds mesh-issues → novox; no bus step; a manifest the change touches fails the module check: modules/mesh-issues/module.json: this …
Every write appends one operation at the issue's last sequence before the
bucket is written; a refused append writes nothing. The start carries what
the bucket holds, refolds what is behind and indexes every issue in memory,
so list answers from memory with a ranked search and a before cursor.
mesh/merge-gate pass: builds netcheck → ace, g14, novox, shanks; no bus step; every machine composes with the change as it did without (4 of 4 compose)
On novox the tool runner runs as the operator's account, and wg show
answered 'Operation not permitted'. Root is the module's concern: the
command runs as given when the process is root, else through sudo -n,
which the sudo module allows the runtime's account, as nftables does.
ace's tunnel leg lost most packets while its own check called the tunnel
healthy, and nothing on the mesh could say which address and port the
hub saw for each peer, which is what tells a home router's port mix-up
from anything else. netcheck_tunnel answers each peer's endpoint, last
handshake and traffic, never a key, and netcheck now says when a
peer's endpoint changes or its handshake goes stale or fresh again.
A machine that backs up nothing read as missing its night after 26 hours up, and a laptop's count, last
written before it slept, read as stale on waking until the next look (novox/hq issue 489).
mesh/merge-gate pass: builds claude-code, mesh-knowledge → ace, g14, novox, shanks; no bus step; every machine composes with the change as it did without (…
A judgement over one bus call's 30 seconds is collected by its ticket from
claude-code, so an eleven-clause record can be reviewed; a proposal out of
review tries is told to the operator, not raised as the module's fault, so
it no longer holds the module's own rollouts.
A merge left out for sharing a module read as containing the stopped merge; the
safety net also held back merges stranded by their walk, which nobody stopped;
a held walk said why only in the log; and a stop of a merge in a batch did not
ask the batch.
A merge walked again into a batch and then stopped by its group stayed in that
batch, which was later let go. Any walk or batch that names a stopped
delivery's merge is now asked to leave it out and is never let go; a refused
leave-out stops the whole walk. A person's stop marks the delivery stopped only
after its walk is, and the answer says what stays on the machines and what
will still carry the change.
A group stop ended the stopped member's walk, and with it every other delivery
the batch put on that walk. It now asks delivery-stop to leave the stopped
merges out, so the others are walked again; a person's stop on a shared walk
first says what ending the walk would take down and acts only on a way.
mesh/delivery covered after its failure: a later walk that contains it was delivered: plan-1791744131193016639 (novox/mesh-catalog at 3fa07682 into main,…
A notice left open kept telling the operator a record was proposed and not in force after it was proposed again
or accepted, and a later version's notice merged into it.
The delivery link's own pattern refused the slash and the @ every delivery id
carries, so no delivery could be linked either. Mirror mesh-delivery's IDOf.
A failed review's try count was already kept with the version, but not when it was asked again, so a 10-minute
pause read as a review lost to a restart; and after the last try only the history and the health said so.
The tracker's own pattern refused the slash a delivery's id carries, so no issue
could track a delivery's condition and the operator's alert kept saying no issue
tracks it. Mirror the controller's key rule instead of a pattern of our own; a
router test proves a slashed key is matched and named end to end.
#205's "place": "." resolved to /var/lib/docker, where the node-engine places
nothing (issue 339), so ace refused the apply and put the module back. The state
directory now says "place": "mesh", <root>/mesh/docker, and a test holds every
directory and path of the module out of the container runtimes' data (novox/hq
issue 452).
The first live review sent research 046 back for a touches link the rules never named and a status proposed
that is the record's state, not the effort's. The rules now list LinkKinds, the one list propose holds a link
to, each with its meaning.
mesh/merge-gate pass: builds mesh-issues, mesh-knowledge, records → novox; no bus step; every machine composes with the change as it did without (4 of 4 co…
mesh/delivery covered: a later merge that contains it was delivered: novox/mesh-catalog@1b7f79f76177 (merged as 73367192 into main, walk plan-17917161183…
mesh/delivery covered: a later merge that contains it was delivered: novox/mesh-catalog@1b7f79f76177 (merged as 73367192 into main, walk plan-17917161183…
mesh/merge-gate pass: builds mesh-issues, mesh-knowledge, records → novox; no bus step; every machine composes with the change as it did without (4 of 4 co…
Once the records leave hq (issue 426), records_decisions would answer every topic empty, reserve would read 0
and hand out 0001 from a counter not started, and import would say it created nothing as a success. reserve now
starts and raises its counters above mesh-knowledge's knowledge_highest, the floor alone starting one (with a
warning) when it is silent and nothing spent with neither.
mesh-delivery says stopped for a pull request closed unmerged too; read as
merged, it was said stuck to the operator for good. Cite ADR 0284 for the
worker rules, which the comments gave as 0283.
The import at hq f0892019 lost reconstructed, amended and a research effort's touches, held rests-on only as a
citation, took eight records' clauses from a numbered list inside a ### subsection, and left the 11 records with
no Decision section clause-less with nothing queued. Each is held now; refresh repairs a record only the import
ever wrote, and never one changed since.
The operator found 13 issues in fixing whose fix was delivered hours before:
their pull requests were linked, not in fixed-by, so their delivery was never
measured, and nothing said an issue left there or a worker gone.
A hung night, a lock held for hours or a lock that cannot be taken left no
failed night behind, so it stayed healthy. The night's process now counts
how long the machine was up since its newest good night, by the monotonic
clock so a sleeping laptop is not blamed, and past 26 hours it is raised.
The retention file is written 0644 and root's instead of read through
sudo: the account reads it and still cannot change it. An unreadable file
names its cause, and the healthy answer says only what was seen.
Each test fails on main: a linked pull request is never measured, nothing
warns of a delivered fix left in fixing, and list has no flag to find either.
The mesh proves which machine called, not which agent on it (hq issue 365), so a review verb could not tell a
reviewer from the author. Each version proposed is now judged by claude-code.claude_code_judge in a separate
call that sees only the record, the records it links and the rules: approve accepts (or asks the operator for a
needs-operator record), changes keeps it proposed with the findings for its author, and a judgement that cannot
be had or read leaves it proposed, says so, and never counts as approval.
A failed night was a journal line only, so a typo in a retention setting
let the repository grow until a disk filled. The night and the measuring
now run as processes the node-engine judges: the night by backup_health,
which is unhealthy for a module not backed up, a night that did not run,
a failed thinning or a missing retention file, and waits naming the
setting when one is refused; the measuring by its unit. A file lock
keeps a night and the tool runtime's now and restore apart.
The retention file is root's alone, so the account reads it through sudo:
until now every night would have refused it as unreadable. Every machine
has the file, so a missing one is refused instead of thinned by 14/8/6.
backed-up says the retention in force and the last thinning per module.
mesh/delivery covered after its failure: a later walk that contains it was delivered: plan-1791711936761655786 (novox/mesh-catalog at 83c64dcd into main,…
A stop is final, so a later merge whose queued walk would carry the modules
now holds it off; a stopped one stays in stalled until a later delivered merge
carries them. A walk counts only modules it sent, and an old walk record is read
again rather than taken as building nothing.
The watch ran as root and wrote its verdict root's and 0600 into the state
directory, which the mesh gives to the account the runtime runs the bundle as;
knowledge_health could not open it, so the module read unhealthy with nothing
wrong. Run watch as that account, as zfs's watch does.
A batch's walk the controller judged to move nothing is done when it is cut,
and a merge it carried stayed published waiting for a start that cannot come.
Its walk's record now settles it: delivered or covered when that walk built its
modules, covered when a later delivered merge carried them (ADR 0270), and
otherwise stopped with why once the cover test says none did.
mesh/merge-gate pass: builds mesh-issues, mesh-knowledge → novox; no bus step; every machine composes with the change as it did without (4 of 4 compose)
mesh/delivery covered: a later merge that contains it was delivered: novox/mesh-catalog@a3f4387e2630 (merged as 300406e0 into main, walk plan-17917125412…
Both pin mesh-sdk/go at 4cf45b5 (mesh-sdk#20 merged). mesh-knowledge's one
reading of the caller, callerOf, now answers stdio.CallerOf, so a call on the
caller-named subject is recorded as its caller's bus user and any other as
unproven, never from the arguments.
The tool runner now takes a call's caller from the subject it arrived on and
hands it to the module through the SDK (stdio.CallerOf); the tracker records
that bus user on what each verb writes instead of unproven. Never from the
arguments, which still refuse any caller, and unproven where the runtime proves
none. Needs mesh-sdk's go/caller.go (pinned at its branch head until merged).
mesh/merge-gate pass: builds new: modules/mesh-knowledge, sent nowhere; no bus step; every machine composes with the change as it did without (4 of 4 compo…
hq's words check never read a decision record, so a correction to any decision
of October 7 to 10 was refused for words it was allowed to write. The links an
accepted record holds are kept by its correction without being judged again.
mesh/merge-gate pass: builds new: modules/mesh-knowledge, sent nowhere; no bus step; every machine composes with the change as it did without (4 of 4 compo…
A revision came from anyone and a review was refused only to the first author,
so a co-writer could approve, and a record proposed before callers are proven
would later be reviewable by its own unproven author. Now a revision is taken
only from the proven author, a reserved number only from whoever reserved it,
and a review is refused from any writer of any version or for any record with
an unproven writer. The rules held records from no date, so a correction to an
older record was refused by rules hq never held it to; each dated rule now holds
from hq's own start date, and a new record cannot be dated before its
reservation. history and the question to ask are bounded like the other answers.
mesh/merge-gate pass: builds new: modules/mesh-knowledge, sent nowhere; no bus step; every machine composes with the change as it did without (4 of 4 compo…
An agent learns what a decision says only by reading whole records. This module
holds every decision record and research effort whole in a database in the store,
with clauses that carry their own status and source, typed links read from both
ends, and every version. It answers ask, get and holds within a stated bound, and
takes a record through reserve, propose (refused on any of the 29 per-record
rules), review by someone other than the author, and acceptance, on the
operator's proven warrant when the record needs it. The 4 whole-set rules make
the module unhealthy when they fail. The import reads hq at a commit through the
records module, never overwrites, and changes nothing when run again. Prose
decisions get drafted clauses that a second judgement must confirm before they
are served.
mesh/delivery covered after its failure: a later walk that contains it was delivered: plan-1791734868297400826 (novox/mesh-catalog at 0a558319 into main,…
The controller pastes a setting into daemon.json as text and the runtime is its
only reader, so a bad size would surface as a runtime that does not start and a
quote could add any key. A step refuses the value first and gates the module.
The defaults are the one hand-set machine's, so it does not change; nothing
restarts the runtime, which takes the keys at its next start.
A number handed out unchecked against hq is a failure that must be loud, so
the answer carries a warning, not only the log. The import already calls
records_status and records_read; invokes now names them.
A new verb, issue-tracker.reserve {kind, for} -> {kind, number}: one counter
per sequence beside the issue counter, advanced the same way, started above
what hq's record holds, a floor setting for open branches and the
reservations kept, and raised when hq holds a higher number. A reserved
number is spent forever, kept with who, when and for what, and said as the
event reserved (hq ADR 0275 rule 8, ADR 0290 rule 5).
A turn given on one carrier's waiting walk let the next member go while
another module of the member before it waited on a walk that could still
fail, and the stop then took the shared walk down. The turn is now shared
only when that one walk carries every module not yet carried, each by a
merge the forge said contains the member's; an unknown answer is asked and
waited for (novox/hq issue 445, re-review of #203).
Under ADR 0276 the merge that carries a superseded member is often folded
into the next member's own walk, which waits for that member's turn: waiting
there waited for itself until the carrier's bound. The members on that walk
now take their turn, as members sharing one walk do. A member that waits says
which merge it waits for and until when, and `stalled` uses that bound; a
carrier still in a batch is bounded by the window and one walk; a member every
module of which was carried is reached, never stopped with no reason
(novox/hq issue 445, review of #203).
Superseded is no failure: the later merge that took its walk over may still
deliver its change, and a covered member counts as delivered (ADR 0270). So the
members after it wait while such a merge is open within its own bound, take
their turn once its delivered walk covers the member, and are stopped, saying
why, only when none can carry it (novox/hq issue 445).
mesh/delivery covered after its failure: a later walk that contains it was delivered: plan-1791682027628755615 (novox/mesh-catalog at 87b0d3a5 into main,…
mesh-delivery read the gate as pending and then wrote "building on", and the gitea module could write the verdict in
between: the forge keeps the newest status, so a pass read pending and a failure hid under "building". The gitea
module now hears the build seat's started itself and puts every write of a head's gate on one chain, marking a
verdict the moment it is heard; a start from before the pending words were last set is not the current check's.
The forge's holder set mesh/merge-gate pending as "the mesh is mapping this head onto its module graph" and nothing
changed it until the verdict, so a ten-minute controller check read as a stalled forge (issue 423). The holder now
says the head waits in the build queue since the local time it was announced; mesh-delivery follows the build
seat's started event and says the machine building it and since when, never over a verdict the forge holds.
The keep-daily, keep-weekly and keep-monthly preferences default to
today's 14, 8 and 6. The night reads them from a file in its state and
refuses anything but a count from 1 to 9999: it then thins nothing and
says which setting is wrong, never handing a raw value to restic.
mesh/delivery covered after its failure: a later walk that contains it was delivered: plan-1791682028101285377 (novox/mesh-catalog at a5b4a757 into main,…
The review asked for the clearing paths to be pinned: a recheck or a merge
clears the line before the forge is read again, and a pull request gone from
the open list is no longer said.
Ready and rejected have no bound, and the watch for unchecked heads skipped
every head with a delivery, so a lost or lagging status write left a pull
request unable to merge with nothing said. Such a head is now said on issue
438's rule: an unanswered line past the table's bound for a check asked and
not answered.
mesh/merge-gate pass: builds mesh-cli → ace, g14, novox, shanks; no bus step; every machine composes with the change as it did without (4 of 4 compose)
Review of #197: a carried superseded member is reached in GroupState but
still stops the members after it in groupTurns (hq issue 445); cite ADR
0239 decision 4, which GroupState extends; shorten the stopped reason so
the member's id survives the forge's 140-character status line. Tests
pin one forge ask per pair and CoveredBy/ContainedIn across a read-back.
A superseded member counts as reached when a later delivered walk on its
trunk carried every module it moves (ADR 0270's covered test); otherwise
the group reads stopped, as groupTurns already treats it. Five groups
read "delivering, 1 of 2 delivered" for days with nothing in progress
(novox/hq issue 442).
Review of #195: the clock restarted on any change in what a head waits on, so a
check answering pushed the line back by the time it took; it now restarts only on
a newer ask. A head whose check started was announced, so its line no longer says
unannounced; pending times are given in UTC like the module's other lines.
The autofs line an armed automount leaves in mountinfo made List say
"mounted" while nothing was mounted. One judgement, Mounted.IsAutomount,
now serves List, the look and the checks before a mount.
The watch for unchecked heads counted any merge-check status as checked, pending
included, so a head left with a pending mesh/merge-gate, or never given
mesh/repo-check, waited unable to merge with nothing saying so. gitea_open_pulls
now gives each head status with its state and time, and the watch says a head
whose required check stays pending or unset past the delivery table's bound for
a check asked and not answered.
systemctl refused the whole call on -.mount, and list answered every mount
without its unit state (novox/hq issue 374). The test fake now refuses a
dash-led argument as systemctl does, so it no longer agrees with the mistake.
The second review found the gap it was asked to look for: everything
after the forge's merge could throw — a forge 5xx, a timeout, a commit
named without parents — and the caller got a bare error that never said
the merge had happened.
From the merge on, a read-back that fails or cannot be done is said
beside merged: true, in read_back; only a proven other head is an error,
and that error says the merge happened. A parent named without its sha
is a failed read-back, never an empty head said to have landed.
A fast-forward's commit is taken as the head only when the forge names
it so. That it does is assumed, not seen on this forge, so another sha
is said in read_back rather than raised as another head having landed.
The method's description lists all five methods and says which two can
be read back. The fast-forward test now gives the merge commit a
different second parent, so reading it as a merge would fail there.
The review of #193 approved it with five nits, all taken.
Seven characters of a sha can be matched by a commit made to match them
in seconds, so an author who force-pushes could have a review merge a
commit nobody read. A head is now named by at least 12 characters, up to
64 for a sha-256 repository.
What landed was read from the forge's record of the pull request, which
is the very thing a race leaves behind. It is now read from the merge
commit: its second parent for a merge, the commit itself for a
fast-forward. A squash or a rebase makes new commits that name no head,
and the answer says it cannot be read back instead of staying silent.
A merge of another head than the one handed over was reported as a
field beside merged: true, which a caller reading only that would miss.
It is now an error that says the merge happened and what landed.
The refusal no longer asserts that the head moved after the review,
which it cannot know; it says the head is not the one named.
mesh-controller#197 was reviewed at 065cfa0d, force-pushed to 3ced96f
at 12:14 UTC, checked green there at 12:22, and merged at 3ced96f at
12:26. The merge tool kept its promise that only a checked head is
merged, but it took no argument for the head the review approved, so it
merged whatever the head was and a commit nobody reviewed landed.
gitea_merge_pull_request now takes `head`, the commit the caller
reviewed (a sha, or at least its first 7 characters), and refuses with
nothing sent when it is missing, is not a commit, or is not the pull
request's head — naming the head it has now. The forge is handed the
same head as head_commit_id, so a push between the read and the merge
is the forge's to refuse. The answer reads back what was merged and says
so loudly if the forge merged another head anyway.
`head` is required, not optional: an optional guard is the shape that
let this happen. A caller that leaves it out is told what to name.
The review's two nits. A write of bytes identical to what is stored was
invisible to a comparison of content, so the test store counts its
writes and the preview's tests read that count: a persist from inside
Preview now fails five of them instead of one. And a router holding an
ask's record and no condition said it held none, rather than saying what
that key is; the branches are the other way round, with a test that
holds nothing else.
The review of #192 found three faults, all verified by running.
The tests did not prove the claim the change leads with: a probe that
sent, queued, recorded and wrote from inside Preview still passed,
because the condition they used was a warning needing nothing, which is
held rather than sent. They now hold an urgent condition, snapshot what
a send would move — the channels, the queue, the history, the refusals,
the stored record and the record held — assert none of it moved, and
tick once more. Every one of those probes now fails the tests.
`cleared` and `silenced` said wording no send would carry: an open
record has no clearing time, so a condition open four hours cleared
"after a moment", and no silence, so a silence read "until Mon 1 Jan
00:17". A clearing is now composed as if it cleared now, which the
answer says; a silence is refused unless the condition is under one.
The preview composed only its own words, while a send reduces a message
whose words the content rule refuses. The escalation walk is now one
function, allowedWords, which `words` calls for a send and the preview
calls alone — so the words read are the words that would leave, and
reading them records no refusal and states nothing on the bus.
withTracking hands back the line it put in, so the preview reports that
line instead of deciding the rule a second time; and a record that is
not a condition, or one carrying an open ask, is refused or says that a
channel's message carries more.
The words a condition reaches the operator with are composed in the
router, and nothing kept them: a send's history row holds a key and an
outcome, never text, and a notification still on the desk is not in the
notifier's history. So a wording could only be proven by waiting for a
send and reading the operator's screen, which is why proving issue 400's
tracking line stalled.
messenger_preview composes what a send would carry for a condition the
router holds open — the same words, the same tracking line, the same
verdict — and says whether the content rule would let them out. It
sends, queues and records nothing.
The operator could not tell from a notification whether its cause was
known, tracked and being fixed, or whether they had to act. The issue
tracker gains the verb tracking, answering for every condition key an
issue links which issues link it and where each stands; the operator
channel's router asks it outside the work and says one line from it in
every condition's message, per hq ADR 0285.
A password holding '=' and a space or comma splits the entry, and the tail
read as name=... was said as "the source <tail>", carrying part of the
secret into the health answer and the operator's conditions (review of
mesh-catalog#190, issue 386).
A password typed into an entry by mistake, in another scheme or split by a
space or a comma, was quoted back in the problem and so reached the status
and the health answer. An entry is now named by its source's name or its
position only; a user with a control character or a second @ is refused
(review of mesh-catalog#190, issue 386).
The live module check refused smb-users: a key naming a user is the
operator's value and may have no default, and a setting without one would
refuse to compose mounts on every node until set. The username now goes in
the source's URL (smb://<user>@<host>/<share>), percent-decoded; a
password there is refused unsaid, and a source without a user waits for
sources (hq ADR 0283, issue 386).
mesh/merge-gate fail: builds mounts → shanks; no bus step; a manifest the change touches fails the module check: modules/mounts/module.json: this manifest …
A configuration without the passwords directory would have read a path
relative to the process's working directory; it is now refused as a fault.
Review of mesh-catalog#190 (issue 386, ADR 0283).
mesh/merge-gate fail: builds mounts → shanks; no bus step; a manifest the change touches fails the module check: modules/mounts/module.json: this manifest …
The one INI secret smb-credentials could not be given through the desk
prompt, which takes one hidden line, and a missing login was said as a
fault: the first-node gate put the build back and the operator was told
nothing was theirs to do. The username is now the setting smb-users and
each password the one-line secret smb-password-<source>; a source missing
either answers waits through mesh-sdk's health answer (hq ADR 0283,
issue 386, issue 379).
The note's by is who acted as the runtime proves it (ADR 0275 §5); the
note's text names both workers. Tests for a resolved issue corrected, link
and set keeping the worker current, and a take-over reason with a secret's
shape. The ready-to-verify note says an issue resolves from fixing.
Two agents could start on one issue without seeing each other, and an issue
could be resolved without anyone having said they started its fix. A move to
diagnosing or fixing now names by, the worker as who@node; get and list show
the worker with since and stale; another worker is refused while it is
current unless take-over says why (noted); a worker untouched for
worker-stale-after (24h) may be taken over from without a reason. Resolved is
reached only from fixing. The moved event carries by, since and caller.
Fixing counts as not closed for the triage and the duplicate check.
hq ADR 0283.
mesh/merge-gate fail: builds mesh-delivery → novox; no bus step; a manifest the change touches fails the module check: mesh-delivery claims mesh-delivery a…
mesh/merge-gate fail: builds mesh-delivery → novox; no bus step; a manifest the change touches fails the module check: mesh-delivery claims mesh-delivery a…
The operator's budget (a leaf module running everywhere within five minutes
of its merge, a core module within ten) cannot be held to without measuring
it. Each delivery now keeps its delivery time, class and phases from the
walk the controller says; 'times' answers the week's median and worst per
class, the deliveries over budget and their longest phase, and one
delivery's breakdown; 'deliveries' is headed by the same line. A phase the
walk could not measure stays unknown, never a zero. Measurement only.
keep wrote the whole delivery on every call, and Reconcile passes every walk
the controller keeps through settleAndKeep every 30 seconds, so the
deliveries bucket was rewritten about once per walk per pass with nothing
changed: a key's revision no longer meant the delivery moved, and a watcher
of the bucket would receive a stream of non-changes.
keep now remembers each delivery as it last wrote it, encoded, and skips the
write when the delivery encodes exactly the same and is not waiting to be
written again. Anything a delivery holds that changes - its state, steps,
walk record, owed effects and their tries - changes its encoding, so every
change is still written; a failed write stays dirty and Flush writes it
again. Events, notes, views and forge statuses are owed and done as before.
A retry of an owed effect still writes its new try count: it is a change.
The test fails before (five passes over unmoved walks wrote five times) and
passes after; a walk that moves is still written and said.
The operator approved measuring from the delivery's own events whether each fix an
issue names reached the machines; delivered is not resolved, and a merged fix that
does not arrive is said on the issue and the operator's channel instead of found.
The operator asked for the judgement that already reads the backlog to fill what an
issuer cannot know, and for dependencies between issues, an issue being workable
only once what it waits for is on the machines. The triage fills only empty values,
adds links and removes none; readiness follows mesh-delivery's own events, said once
per spell, through a duplicate to its original, and loud when a delivery cannot be asked.
mesh/merge-gate pass: builds dnsmasq, docker, netcheck → ace, g14, novox, shanks; no bus step; 2 wait(s) for a person; every machine composes with the chan…
mesh/merge-gate pass: builds claude-code, mesh-issues → ace, g14, novox, shanks; no bus step; every machine composes with the change as it did without (4 o…
An independent review found a pass could starve new issues' checks, two writers
could apply one verdict twice, a judged issue resolved by a person could still be
linked, and the dry run's answer raced its pass.
mesh/merge-gate pass: builds claude-code, mesh-issues → ace, g14, novox, shanks; no bus step; every machine composes with the change as it did without (4 o…
The operator asked for duplicate issues to be consolidated as they are opened.
The tracker consumes its own opened event and asks the agent on its node, through
claude-code's new claude_code_judge, whether an older issue not resolved asks the
same question; it closes only the newer issue of a certain duplicate, keeping its
report on the older one, and fails loudly when it cannot judge. groom runs the
same judgement over the backlog as a dry run, applied on confirm.
dunst.started says the notifier's bundle started again: the notifications stay on screen and only the
answers they waited for are lost. Forgetting their ids made the router's message a second popup beside a
dead one. The desk now keeps the ids and says answers-lost-at; the router's message replaces each
notification in place.
Whether a message is said again is now its own (last said before the answers were lost), so a router
that started after the loss still says it and a repeated standing says nothing. It is sent to the desk
directly, not through the outbox, whose failure path moves a warning to Telegram; a failed send is tried
again at the next standing.
On 2026-10-10 the desk's notifier started again on two machines and the desk showed, from its own
memory, nine old popups: resolved, silenced and cancelled ones among them. Its memory held every handle
ever shown with buttons, and the router's edits (silenced, resolved) had reached only the newest handle
of each problem, so the rest stood as if open.
A channel's memory serves only to replace or edit a message in place. A notifier that started again lost
its screen: the desk forgets that machine's notification ids and says in its standing since when the
screen is new (shows-since). The router, which alone knows what is still open and unanswered, says that
again on that kind: one message per problem under its handle with the words and answers of now, a
silenced one edited to say so with no answers and no urgency, nothing that ended. Each move of
shows-since is acted on once, and waits while the desk cannot send rather than going to Telegram.
Telegram re-shows nothing on a restart (its messages stay in the chat) and says no shows-since.
An edit that offers nothing and reaches no screen its message is on is taken with nothing to change;
one that offers answers is said as failed, so the router says it anew.
mesh/delivery covered after its failure: a later walk that contains it was delivered: plan-1791663540935383995 (novox/mesh-catalog at 7534167f into main,…
The secret smb-credentials was an INI file alone, which the desk's one-line
prompt (secret ask, hq ADR 0277) cannot type: on the desktop the Pi's source
said "line 1 is not key=value inside a [section]". A login is now also one line,
`[recalbox] username=root password=…`, password last and to the end of the
line so it may hold spaces and `=`; the file form still reads, and a file may
mix both. A line the module cannot read is refused with the form it expected,
never with what was typed, and a login without a username is not answered. The
source's "login is not given" says the desk verb and the line to type.
The controller's secret-ask verb opens the launcher's hidden prompt from anywhere on the mesh, so the
operator must know whose ask they are typing into. The prompt takes an optional asked_by, the
controller's word on the bus's caller held to a name's characters and spaces, and says it; a sentence,
markup or a line break is refused, so the prompt's words stay the launcher's.
mesh/merge-gate pass: builds desk-channel, messenger, telegram → novox; no bus step; every machine composes with the change as it did without (4 of 4 compo…
mesh/delivery superseded: its walk was superseded and names no walk that took it over: a walk that took it over answers its merge, and it follows that wa…
Seen at 14:40: machine.shanks.units was silenced for 7 days with a reason,
and its critical popup stood on the desk as "Still open" while the
router logged nothing. A silence now edits the message where it is shown
to "Silenced until <day>: <headline>" with the reason, no answers and no
urgency, so the desk's popup times out; nothing more is said while it
lasts. When the silence ends and the condition is still open, the message
is edited back to its open words. Resolved while silenced, it is said only
where an edit notifies nobody. The silence is taken from the controller's
event and from the 10-min reading of the open set alike.
mesh/merge-gate pass: builds desk-channel, messenger, telegram → novox; no bus step; every machine composes with the change as it did without (4 of 4 compo…
- A comeback before the clearing was said rings when it is urgent now, and
says the reminder or escalation that was waiting when it cleared.
- Comebacks never said are counted apart, so "again, Nth time" counts only
what the operator was told; an urgent comeback after a said clearing rings
again in place of its message.
- telegram sends the new message before deleting the earlier one, so a
failed send leaves the operator a message.
- desk-channel keeps a machine's notification on record when its replace
fails, and says a message shown nowhere as failed.
- A store that cannot be read sends a new message instead of holding it.
- Tests for each, and for a reminder in a burst and a clearing the channel
could not take.
The operator stopped believing the mesh's notifications: one problem stood
twice on the desk (a reminder beside the message it followed, an escalation
then a reminder 31 s later), a unit failing again at each apply was said
again 22 times, and a question's buttons outlived it on a reminder the
ask's end never reached.
- messenger: a reminder or escalation is said in place of the problem's
message (Replace under its handle), alone, never as a digest line; a
problem first said in a digest gets its own message next time. A
reminder counts from when it was last said. A clearing is said once it
stays cleared 2 min (10 min once it came back); back before that, the
message continues without a word. Cleared before ever said and back:
said. An ask's expiry is said with its day.
- desk-channel: a send or show under a handle already shown replaces its
notification by the notifier's replace id.
- telegram: the same deletes the earlier message and sends the new one,
ringing as its kind rings; one it cannot delete loses its buttons.
The review found that a double tap while the phone's spinner turned would be
answered 'That answer was not taken', reading as if the approval failed; it is
now 'You already chose Approve at 13:30: recorded.' and does nothing more. The
fallback toast no longer promises a change that may not come (a stranger's
tap), the wait outlasts the router's 10 s wait for the controller's root word,
the channel's own refusals stay in the chat too, and the store's error words
stay in the journal.
At 13:30 on 2026-10-10 the operator approved two questions on Telegram and saw
nothing happen: the channel acknowledged each tap with no words, before the
router had judged it, and the edit that removed the buttons kept the question's
title, adding only a line under it that rings nothing.
The router now answers a chosen option in reply to the tap, once the ask's end
is recorded and before the question is edited: "You chose Approve: recorded,
and claude-code was told." (or, when the bus did not take the warrant, that it
is told when it does). A refusal already replied with its reason. Telegram
holds a tap's acknowledgement for the router's word and shows it as the tap's
own toast; a word with a body (a refusal's reason, Details) also stays in the
chat. Without a word in 8 seconds the toast claims nothing, and the late word
goes to the chat. A secret is never a toast.
An ask's own message, once answered, carries its outcome as its title:
"You chose Approve at 13:30 on Telegram · <headline>", then whether the asker
was told. The asks protocol carries no result back from the asker, so nothing
more is claimed of what it did.
The review found that each try at an unready channel composed its held messages and so made a Details
ticket for each, and enough tries pushed out the tickets of messages already shown.
A newer message about the same problem (an escalation, a reminder, a re-send) replaced the Details
ticket of the one before it on that channel kind, so tapping Details on any but the newest was refused
"it belongs to no open message". Each message now keeps a ticket of its own, used once, at most
MostDetails per record and kind and none older than TokenFor; one pushed out is refused in words that
send the operator to the newest message. A change of the condition's actions no longer drops Details,
which names no action.
mesh/merge-gate pass: builds claude-code → ace, g14, novox, shanks; no bus step; every machine composes with the change as it did without (4 of 4 compose)
A refusal heard on the decided subject is kept as heard, and a record of a person's choice found later is
still taken, so an ask the router's queue offered again is not lost in silence. A heard end must match the
proposal's ask; the wait stops reading at its deadline; the heard ends are forgotten when Propose returns.
Two catalogue merges 18 s apart were walked by one walk at the later commit,
which superseded the first; this owner tied that walk to the dunst delivery
the supersession had ended and to no other, so nothing gave it its word and
claude-code was held for a person with a reason naming neither cause.
A delivery now follows the newest walk that answers its merge: one naming it
(delivery.merges, the controller's batches), its own commit, the walk that
took its own over (taken_over_by, or the superseded note), or, from a
controller that names no merges, an open walk of its branch made for a later
merge, moving all it moves, whose commit the forge says contains it. Only its
own walk's record moves an ended delivery. The word is asked once per walk,
when every delivery it carries is at its turn, a group's members on one walk
together. An earlier merge carried by a later commit ends covered with the
walk, and waits for a walk of its own when it fails. A merge no walk answers
is held with the true reason and listed at once; a waiting walk no live
delivery holds is listed as unheld; a merge in an assembling or queued batch
is shown, not held. Replay R362 fails before.
mesh/merge-gate pass: builds claude-code → ace, g14, novox, shanks; no bus step; every machine composes with the change as it did without (4 of 4 compose)
A module that only asks is issued no channel kinds, so the empty list was read as 'no channel proves its
sender' and every proposal was refused on the live mesh with Telegram linked. The router judges; Propose
waits for its word, and a refusal heard on the decided subject discards the proposal on every node.
mesh/merge-gate pass: builds new: modules/mesh-issues, sent nowhere; no bus step; every machine composes with the change as it did without (4 of 4 compose)
From the review: a taken key that cannot be read back, or a write never
confirmed, now tells the caller to get the number before opening again; an
owner entry naming a module is owned only by that module registered; link
targets get the secret guard; line separators are refused in one-line fields;
the acceptance test now proves the counter's own updates were contended.
mesh/merge-gate pass: builds new: modules/mesh-issues, sent nowhere; no bus step; every machine composes with the change as it did without (4 of 4 compose)
The seat issue-tracker, held once for the mesh: open, get, list, move, note
and link. The counter is advanced by the SDK's compare-and-set Increment
(go/v0.1.12) and started once at the setting first-number (365, after hq's
highest), never below an issue kept. The cycle's issue rules are refused at
the write; every change is an event. The runtime does not yet say who called,
so every change is recorded as by "unproven" and no argument can name a caller.
mesh/merge-gate pass: builds claude-code → ace, g14, novox, shanks; no bus step; every machine composes with the change as it did without (4 of 4 compose)
mesh/merge-gate fail: builds claude-code → ace, g14, novox, shanks; no bus step; a manifest the change touches fails the module check: claude-code uses the…
An @ after any whitespace or punctuation, invisible and control characters, and top headings are refused;
every node judges a proposal record as the module would have made it before applying an approval, never
applies one older than its section's last change, never takes an unreadable record for an empty one, and
keys a node section by its nodes. The controller records the warrant from the router's word alone.
mesh/merge-gate fail: builds claude-code → ace, g14, novox, shanks; no bus step; a manifest the change touches fails the module check: claude-code uses the…
Shared instructions were the terminal's alone since hq issue 340, because the tool runner cannot tell an
agent from the operator. The operator channel now proves who answered, so an agent may propose a section;
each node proves the router's warrant against the exact text itself and keeps what it approved in its own
file, so nothing written to the bus becomes active.
mesh/merge-gate pass: builds build-agent, gitea, route-proxy → ace, g14, novox, shanks; no bus step; every machine composes with the change as it did witho…
Every merge to the controller's repository rebuilt the route proxy and the
build seat's holder, because their builds read that repository whole. Named,
the build seat hands each only its program's build source and the planner
moves each only when that changes. A rename was said under its new name
alone, so a file moved out of a build source changed nothing the planner
could see.
mesh/delivery covered: a later merge that contains it was delivered: novox/mesh-catalog@e4edab3fd0eb (merged as c41c32cb into main, walk plan-17916298010…
mesh/merge-gate pass: builds claude-code → ace, g14, novox, shanks; no bus step; every machine composes with the change as it did without (4 of 4 compose)
Typing claude on the control node should start the session as the agent's own account, not as the
operator. The controller fills no machine fact into shell code, so the account agents run as is
written to /etc/claude-code/agent-account and the zsh code adds alias claude=claude-agent only when
that file names an account other than the shell's own. Where agents run as the operator's account,
nothing is added. claude-agent keeps running the real claude: exec sees no alias.
mesh/merge-gate fail: builds dunst → g14, shanks; no bus step; 2 problem(s) the change brings; the first: a machine (f35): nothing could be sent to it with…
The actions sat on the middle click, which the operator's touchpads cannot give, and the context
menu named dmenu, which dunst's PATH does not hold. The left click is now do_action, close_current
(one action runs, several open the menu, none closes), the menu is rofi, declared as a package,
Super+N opens it from the keyboard through i3's contribution, and send refuses the token default,
which dunst would run at a tap without the menu.
mesh/merge-gate pass: builds claude-code, claude-licence-manager, restic, sudo → ace, g14, novox, shanks; no bus step; every machine composes with the chan…
mesh/repo-check pass: THE CHANGE ALTERS ITS OWN CHECK (merge-check.sh): main's version judged it; the change's judges the pull requests after it merges; it…
The managed settings apply to every account's claude, so a key-helper script under the
agent's home would run as the operator, and through sudo as root: the command is now
carried whole in the root-owned managed settings and reads the key only as the agent.
The helper reads only a regular file, bounded, without blocking, and a request it does
not answer fails in time instead of holding every later one; a directory entry it
answers that is not a name is refused. The guard's record is matched on its binary in
the query, a half-named agent account is not taken as the operator's, the kept API key
stays until the agent's home holds one, restic refuses to restore beside a directory
others may write, and the agent-home check cleans a path before judging it.
mesh/merge-gate pass: builds claude-code, claude-licence-manager, restic, sudo → ace, g14, novox, shanks; no bus step; every machine composes with the chan…
mesh/repo-check pass: THE CHANGE ALTERS ITS OWN CHECK (merge-check.sh): main's version judged it; the change's judges the pull requests after it merges; it…
mesh/delivery covered after its failure: a later walk that contains it was delivered: plan-1792148283013560219 (novox/mesh-catalog at 3db85a0d into main,…
A condition told the operator to type the node-engine's binary at the machine, and nothing held the catalogue's texts to ADR 0272. The words check gains retired-lines — the controller's and the node-engine's commands, copied from hq's words.py --list lines, which compares it — and refuses one in any string a module's tools can show or any manifest description. The vault's and the route adapter's texts now say the nox line.
The review of 2026-10-09 (M4): the prompt and its message were the caller's free text, so whoever could
ask the verb could put any words before the operator. The verb now takes the module, the secret and the
machine, checked as names, and writes the prompt here, saying the mesh's controller asks — which the bus
makes true, denying the verb to every other principal (mesh-controller #156). prompt and message are refused.
The review of 2026-10-09 (H4). The desk ran in the operator's runtime, where agents run; it now runs as
the account desk-channel, as telegram does. A link's code is shown only on a named machine whose notifier
said in the last five minutes that it is not paused, and is sent secret, so the notifier keeps it out of
the history every caller of the verbs reads; while no such machine exists the desk says it cannot send,
and the router starts no link.
The review of 2026-10-09 (H4): a link's code shown on the desk stayed in dunst's history, which
node-notifier.history serves to every caller of the mesh's verbs, agents among them; and a code sent to a
paused notifier is held and shown later, to whoever is then at the screen. send now takes secret (transient,
category mesh.secret) and transient; the dunstrc's rules keep both out of the history, and history leaves out
any that got there. The pause, read from dunst itself, is said as the event paused at start, on change and
every two minutes, for the desk to read.
The review of 2026-10-09 (M4): another holder of the token can change a message's words and leave its
buttons, so the operator taps Release under words that say something else. Each message's words are now
kept as a digest, per chat and message, and a tap on a message whose words differ, or that this channel
never showed, is refused on the phone and never said to the router.
The README says, as the operator directed (H5), that an approval from Telegram is meaningful only where
agents run under their own account without sudo, without the operator's X display, and with Telegram
Desktop's data readable by the operator alone; until then, approve from the phone.
mesh/merge-gate pass: builds messenger, sudo → ace, g14, novox, shanks; no bus step; every machine composes with the change as it did without (4 of 4 compo…
The build seat runs the suite as root, where every file the test makes is root's, so "a file not made by
root" was root's and unlinked (the forge's repo-check of 2026-10-09). The owner is now faked as another
account for the refusal and as root for the act; the test fails as before under unshare -r without this.
A word kept from a moment ago, not free, refused an ask taken just after root became free again, and
the asker keeps a refusal until something changes. Taking an ask asks the controller fresh, still
before the work is taken (the lab's proof found it).
N1-router: another account's /start offered the desk an unlink of the operator's account in its first
hour, and a desk click, which a program on the desk machine can forge, unlinked it and barred it for a
day, again and again. An unlink at the desk in the first hour now bars nothing: any account, the
operator's too, links again at once with a code from the desk; the day's bar stays for an account
removed by its own approval or by root, and root's allow-link lifts any bar. The words say which.
Lows: the controller's root-free word is asked before the work is taken (it waited up to 10 s inside
it) and the answer is judged with that fresh word; inside the work only a recent word is read, and a
failed ask drops the old one. An answer whose effect the content rule withholds is not offered.
mesh-sdk moves to 16984aa. Each test failed before its fix.
The review of 2026-10-09: the router read the absence of an agent-root condition as a pass, and its minute
hung on a break-glass file, taking the work it already held. An approval now rests on the controller's
verb root-free, asked live for the router's machine and the channel's, failing closed on every other
answer, and a stale word on the kinds promises no verified sender. Break-glass runs in the minute, says
on every channel what root did, and ends the router's own open questions. An ask offering an approval is
answered only where it could be approved; a message says who asks, what each answer does and what expiry
does. The router's own asks bind the account they are about; a link is checked against one per kind as it
is made; an unlinked account does not come back by a code within a day; a fresh link is unlinked from the
desk; no link begins while the desk cannot show its code, and a code is marked secret on every path. Links
are bounded per account in the router's state; a warrant's message id cannot be foreseen. Moves to
mesh-sdk 76902998 (canonical digests).
The review of 2026-10-09: an unknown account and sudo that could not be asked were answered not root, and
only NOPASSWD on ALL counted, though NOPASSWD on systemctl, tee, bash or pacman reaches root as surely,
and a password the person typed is kept by sudo for minutes. Any rule now counts, as the node-engine's
own judge of an agent account does (ADR 0266); what cannot be judged fails the answer. The bundle built
in the module's directory is removed and ignored.
The lab's proof showed an ask whose approval was held back (an agent could become root where the
channel runs) folded into a burst's digest with a condition, and its one remaining answer, Decline,
lost with it. A message carrying an ask's answer now always goes alone.
The lab's proof showed a stranger's refused tap said as "That answer was not taken" on the operator's
own chat (the router sends to the linked account), where it reads as the operator's own answer refused.
It now says which account tried, that nothing was done, and what it may mean.
A warrant now carries the digest of the ask exactly as the router took it, so its asker acts only for
the ask the person was shown, with what each option binds. The router refuses an ask whose approving
option binds nothing (the SDK's Check). Its own unlink and relink asks bind the act they perform, and
it performs that act only on a warrant for that ask. mesh-sdk moves to d4077b4.
A build the controller held back for a person reaches its machines by a backlog walk on the
operator's word, which credits the delivery that published the build one step per machine and
never the walk's own "every machine running it". Its last record keeps no gate, only the machines
it is done on, so the credited steps stayed judging for ever, and `covered` read only the aggregate
steps, so the held merge had no end but a person's wrong act (the catalogue's #125, held 28 hours
after hq ADR 0270 went live).
A finished backlog walk now passes the steps it carried on the machines it is done on; `carries`
counts a module as sent everywhere when a passed step stands on every machine the held delivery's
plan sends it to; and a walk that carried a step still judging is asked for by its id, since one
that ended before the holder's restart may be older than the ended walks the controller lists
back. Three tests fail before and pass after.
mesh/merge-gate pass: builds mesh-cli → ace, g14, novox, shanks; no bus step; every machine composes with the change as it did without (4 of 4 compose)
So a secret reaches the controller's terminal without standing on a command line:
mesh-cli secret accept <node> <module> <name> --from - < file. A terminal is never read, more than the
bound is refused before anything is asked, and nothing carried is printed. mesh-sdk moves to f047d0a.
mesh/merge-gate pass: builds dnsmasq, docker, netcheck → ace, g14, novox, shanks; no bus step; 2 wait(s) for a person; every machine composes with the chan…
- The dnsmasq watch restarts only a unit that runs, without waiting for
its start; a unit stopped, starting or failed is held and said (C2).
- dnsmasq_resolve asks only an address the resolver listens on (C3).
- Tests for the watch loop keeping its budget, a watch that wrote nothing
or stopped looking, and a unit not running (C4).
- A restart is remembered as said only once emitted, restarts made before
the bundle started are said, and whether one helped is said (C5).
- docker_events reads the runtime's whole buffer: complete when it is not
full or reaches past the window, never a quiet machine read as lost (C6).
- netcheck: refused read with errors.Is; tests for addresses and ports,
and for a container named as docker knows it.
bind-dynamic closes every listener whose interface a re-read of the
machine's addresses did not find, and a re-read during a bridge teardown
fails (dnsmasq 2.93 network.c:832-875, reproduced in a network namespace:
53 releases in 120 parallel teardowns). bind-interfaces never closes a
listener; a drop-in makes the unit wait for the private address and retry
a start that found it missing.
On 2026-10-09 the control node's route-proxy failed every lookup against 10.10.0.1 alone for twenty
minutes while the machine's file listed both mesh resolvers; its copy lives under the runtime's
root-only directory and no tool read it. docker_resolv_conf reads /etc/resolv.conf as the container
sees it through the runtime's archive (docker cp, no exec), parses it, and says whether it lists the
same resolvers as the machine.
During the 2026-10-09 outage the control node's resolver stayed active while it no longer listened on
its private address, and nothing on the mesh could ask that address a question or show which sockets
were bound. netcheck_dns takes server (and tcp) to ask one resolver alone and say its response code;
netcheck_listening reads the kernel's socket tables, as ss -lunt does, with the owning program where
the tool's account may see it and a sentence where it may not.
docker_events asked for three hours across the 2026-10-09 resolver outage on the control node answered
zero events: the runtime keeps only its last ~256 events in memory, and health-check execs filled them
with the last ten minutes. The answer now carries held, oldest_held and complete, and a history line
when the window reaches further back than what the runtime still holds. The type filter moves into the
tool, since the history's start is only known from all of it.
On 2026-10-09 the control node's dnsmasq stayed active and running, its
journal silent, and refused every question on its private address for
twenty minutes; nothing in the module looked past the unit.
The module is now Go (its TypeScript tools were never served): a root
watch asks the machine's own name on every listen address over UDP and
TCP every 15s, restarts the unit after two failing looks (2 min apart,
3 an hour), and says whether that helped. dnsmasq_health is the watch
process's declared check, so a restart that did not help becomes the
module's unhealthy condition; the service is judged by its unit.
mesh/merge-gate pass: builds nfs-server, sent nowhere; no bus step; 1 wait(s) for a person; every machine composes with the change as it did without (4 of …
- mesh-cli pins the SDK at its tag now that go/v0.1.9 exists, instead of a
pseudo-version of the branch it was cut from.
- mesh-delivery's stalled line for a head the forge never announced
carries the pull request's number, which the controller's condition
names in its headline.
After merging #159 (issue 347): every /repos/<owner>/<name> the client builds
now takes its owner and name through one helper, whether they came from a
caller or from the forge's answer; the open-pulls fixture's shas are commits.
The confirmation review of 2026-10-09 found the dedicated tools building paths
from caller text: a branch named .. became DELETE /repos/o/r/ (the repository),
a file path ../../../../version reached /api/v1/version, and a sha with
?sudo= set a status as any user. Each caller-supplied segment now goes through
one helper refusing empty, . and .. and / ? # %; a sha must be a commit; and
the client judges every path once more as it sends it: no dot or empty
segment, encoded or not, and only the allow-listed query keys.
The confirmation review asked for the delivery domain's own vocabulary
instead of a provider standing. A gitea health check would need a `tools`
list (turning gitea's tools into seat verbs), a check of another kind
beside it, and would hold the forge's own deliveries at the first-node
gate; so the delivery's owner says it, with the word it already has.
- gitea_open_pulls (gitea): every open pull request from the forge itself,
with what its base requires and the statuses its head has.
- mesh-delivery asks it every five minutes and keeps its own first-seen
time per head (the forge's update time moves on every comment). A head
on a base requiring mesh/merge-gate OR mesh/repo-check (hq requires only
the second) with neither for ten minutes is a `stalled` line in state
`unannounced`, which the controller raises as delivery.<id>.stalled (D14);
H2 has nothing to close. Once announced it is a delivery of its own.
- The gitea-side watcher and its provider standing are gone.
The confirmation review of 2026-10-09 measured GET /user?sudo=… answer the
operator's own account: the query went to the forge unjudged, and the tools'
token is the site administrator's, so any caller could act as any user.
gitea_api now sends only allow-listed query keys and refuses sudo in any case,
and a review that approves or requests changes is the operator's.
The announcer read only the repositories whose update time moved since its
last pass, less a minute. Opening, reopening or retargeting a pull request
does not move a repository's update time, so a pull request opened more
than a minute after its push was never announced, never checked, and
nothing said so (mesh-sdk #14, 2026-10-09).
- Each pass also looks at every repository holding a pull request updated
since the floor, by the forge's own search; every ten minutes, or when the
search fails, at every repository.
- Every five minutes the forge is asked for every open pull request; one on
a base requiring mesh/merge-gate whose head has had none for ten minutes
is said as the forge failing that pull request's merge check (a provider
standing, ADR 0224), which the controller keeps as a condition, and is
recovered when a check arrives or it closes.
- The client refuses a pull request or issue number that is not one before
asking the forge (a misnamed tool argument sent pulls/NaN).
mesh/merge-gate pass: builds claude-code, route-adapter, new: modules/mesh-cli → ace, g14, novox, shanks; no bus step; every machine composes with the chan…
A line whose answer did not come in time is said as possibly run, with
its call, never as nothing having run. A test holds the module to one
archive resource unpacking one compiled Go bundle into /usr/local/bin.
The README says the MESH_VERB refusal is a courtesy and the controller is
the gate. SDK pinned at the go/cli branch's newest commit until go/v0.1.9.
user-dirs.dirs is shell assignments meant to be sourced by a shell, and a setting is pasted into it as
it is, so a folder of $HOME/x"; <command>; X=" set through the verb would add a command of the
caller's. With the file trusted again, the three folders are set at the controller's terminal alone
until settings are typed and quoted at render.
mesh/merge-gate pass: builds new: modules/betterbird, sent nowhere; no bus step; every machine composes with the change as it did without (4 of 4 compose)
A composer window whose class is not Betterbird's own was not seen, so a restart could lose its message.
Each step of a restart now has a bound short enough that all of them together, taken in full, stay
under the runtime's 30 s: 3 s for the window tree, 8 s to quit, 3 and 5 s to start, 4 s to see it.
A setting is pasted into picom.conf as it is, so a backend of xrender"; write-pid-path = "… adds an
option of the caller's: picom would then overwrite any file the operator can write. With the file
trusted, both settings are set at the controller's terminal alone until settings are typed and quoted
(hq issue 346).
Follow-ups of the review of 2026-10-09: the escape hatch sent the raw path it
had judged in normalised form; a pull request's base could be moved through
it; the merge tool would merge into a branch no rule guards; a rule named for
a branch could undercut the glob rule guarding it; and the delete tool took
any unguarded repository. It now takes only a repository of the tools' own
user with no protection and no commit in 30 days, named in full.
A window mapped but left unpainted, or stale after a workspace switch, is a known failure of picom's glx
backend with damage tracking on some drivers, and the first suspect for the empty calendar reminder on
the workstation. Research 037 already found the backend to depend on the machine's driver. Both become
preferences whose defaults are today's (glx, use-damage true), so nothing changes until a machine is
given xrender or use-damage false to test it.
A window that is unmapped or on a hidden workspace was cut from the screen anyway, which shows what is
there instead of it. An inline image of up to 4 MiB could not cross the bus, whose largest message is
1 MiB; it is now held to 700 KiB and halved in size until it fits, saying so.
mesh/merge-gate pass: builds new: modules/betterbird, sent nowhere; no bus step; every machine composes with the change as it did without (4 of 4 compose)
A restart ended Betterbird with whatever a composer held; now it refuses while a Msgcompose window is
open unless discard_unsent says the operator agreed, and answers within the runtime's 30 s. A
preference's value could carry an address, a user%40domain@host URI, a host or the home; those are
masked, recent-folder, account and cloud-file preferences and every add-on preference but the theme's
id are withheld, and a request for everything is refused. A log kept the token after Bearer or Basic.
The reminder test's Betterbird counted as the operator's and left its profile behind when it ended on
its own; it is now left out everywhere and its unit removes the profile when it stops.
mesh/merge-gate pass: builds claude-code, route-adapter, new: modules/mesh-cli → ace, g14, novox, shanks; no bus step; every machine composes with the chan…
The records send the operator to "the controller's terminal", and on the
control-node nothing answered (hq issue 343). mesh-cli is a module of its
own, assigned to every node and built from its own directory: its program
is unpacked into /usr/local/bin by an archive resource (one file, no link,
no copy), and its zsh contribution makes nox its alias. It holds no
credential: it asks the node-engine on its machine over the SDK's go/cli
socket, and prints the controller's answer with its exit.
The agent module's refusals and README, and the route adapter's README, now
name `mesh-cli settings set …` on the control-node instead of a
mesh-controller command nobody could type; a settings file is passed inline,
since a path would be read where the controller runs.
A reminder window that showed as transparent and empty could not be looked at: the window manager's
list leaves dialogs out and says nothing a compositor goes by, and xwd -id refused a window of a
32-bit visual with BadColor (X_QueryColors). xorg_windows answers every window with its class, and
on request its depth, type, state, role and own opacity; a window's screenshot is now cut from the
whole screen as the compositor shows it, and inline answers the image for a session that cannot
read the machine's files. Declares xorg-xwininfo, which both now use.
mesh/merge-gate pass: builds new: modules/betterbird, sent nowhere; no bus step; every machine composes with the change as it did without (4 of 4 compose)
A calendar reminder showed as a transparent, empty window on shanks and no tool could say which
window it was or which preferences Betterbird ran with. The module owns one file, the user.js its
setting prefs fills, which its tools place whole into the profile Betterbird uses and never over an
operator's own; it reads named preferences without credentials or accounts, restarts Betterbird under
the user manager so its output reaches the journal, checks what it promises, and shows a reminder on
demand from a throw-away profile so a fix can be proved without the operator's calendar. No package:
betterbird-bin is the AUR's.
gitea_api took any endpoint with a token minted for the forge's
administrator with write:admin: an agent could drop a branch's protection
or its required checks, set the merge check's status, write to main or
merge its own pull request (novox/hq ADR 0266 §7, the review of
2026-10-09). gitea_api now reads anything but the administration and
writes only what an author writes; protection is only strengthened; a
merge needs every required status a success and merges only that head; a
guarded repository is not deleted; and the token is minted without
write:admin, a kept one re-minted at the next start.
mesh/repo-check pass: THE CHANGE ALTERS ITS OWN CHECK (merge-check.sh): main's version judged it; the change's judges the pull requests after it merges; it…
The controller fills ${machine:agent-home} in a user's home now, so an agent
account the operator names with a home of its own is created there, where the
module places its ~/.claude (hq ADR 0266). An existing account is never moved.
The node-engine writes those as root after checking the path, and the agent
account owns its home and could swap a link in between (hq ADR 0266). No
module does it today; this keeps a future one from opening the gap.
An agent the operator starts as the agent account through sudo (hq ADR 0266)
would otherwise share the operator's terminal and could push keystrokes into
their shell after it ends. sudo still runs without a terminal for the mesh's
own tools, so nothing they do changes.
The control node is to run its agents under an account that cannot become root
(hq ADR 0266). Declare it as a user the node-engine judges root: never, put the
agent's directory in its home, and place claude-agent so the operator starts a
session there through their own sudo. Where no agent account is named, both
name the operator's account, as before.
mesh/merge-gate pass: builds claude-code, claude-licence-manager, restic → ace, g14, novox, shanks; no bus step; every machine composes with the change as …
The check looked for "rt" in a random base64 ciphertext, which holds it
about one time in seventy, and failed mesh-catalog #141's repo check by
chance.
Root writes a restore beside the live data. In a directory another account
controls, such as the agent's home (hq ADR 0266), that account could put a
symbolic link where root writes and have root write anywhere; such a restore
now goes to the module's state.
On the control node agents are to run as an account that cannot become root
(hq ADR 0266). The runtime keeps running as the operator, so the token, the
identity and the home-scope items go to the agent's home through a helper
started as that account; the licence and the account file follow the move,
what was placed in the operator's home goes once the agent holds the licence,
and the guard refuses every tool of a session running as anyone else,
recording to the journal. An API key is taken only from the agent's home and
only when it is one, here and in the licence manager, so the tool cannot carry
off the runtime's credential. Unchanged where no agent account is named.
mesh/repo-check pass: THE CHANGE ALTERS ITS OWN CHECK (merge-check.sh): main's version judged it; the change's judges the pull requests after it merges; it…
A change to a module's tests moves no build (hq ADR 0267), so the planner's modules leave it out; the
TypeScript part also reads the modules the changed files are in.
It had not been run before a merge: its expected list lacked gitea_commit_statuses since that tool came,
and now gitea_contains. The catalogue's TypeScript check, merged in from #151, runs it.
mesh/repo-check pass: THE CHANGE ALTERS ITS OWN CHECK (merge-check.sh): main's version judged it; the change's judges the pull requests after it merges; it…
mesh/repo-check pass: THE CHANGE ALTERS ITS OWN CHECK (merge-check.sh): main's version judged it; the change's judges the pull requests after it merges; it…
The catalogue's check ran only Go. A TypeScript module's change, the forge's among them, ran nothing of
its own before it merged, and a type error would first have shown in its walk. A second part, declared
with mesh-check-also and run in the TypeScript toolchain, type-checks each touched TypeScript module with
the compiler and flags a build uses, against the SDK and Node types the toolchain holds, and runs its
tests as its own npm test does. A module that needs a package the toolchain lacks, or has no tests, is
said by name, never passed. The media catalogue's twin is hq issue 311.
Merge order on a protected trunk was the whole proof; the review asked for ancestry. The gitea module
gains gitea_contains, a read-only compare of two commits (total_commits 0 is contained), and mesh-delivery
asks it once per pair, off the lock like its plan ask, keeping the answer on the delivery. A forge that
cannot say leaves the merge held, not covered. Merge order stays the cheap first filter.
A later delivery now carries a module only when its own check passed (so a merge held without a passing
check is closed only by one that passed), its walk is of its own merge commit on the same trunk, and its
steps say the module reached every machine running it, not one. The words say what is known: a later
merge that contains it was delivered.
A held delivery asked a person a question with no right answer: four hq merges that build nothing,
merges adding a module the controller sends nowhere, controller merges whose change a later walk of
the same trunk already delivered, and a group member held because its composed check had not yet
answered. Only stop could end them, which marks a delivered change as an error on its pull request.
- A merge whose plan moves no module (nothing, or only a new module sent nowhere) is delivered at once.
- A new final state, covered: every module a merge moves was sent by a finished walk of a later merge
into the same trunk, which contains it. Shown as success on the pull request; a group counts it as
delivered.
- A group member waits published while its composed check is asked and unanswered, and a member held
for it is lifted back to published once it passes for the heads that merged.
Read again each tick, so the deliveries already held end by themselves after the rollout.
Third review of #147. The run writes as root into /var/lib/zfs-scrub (0700, root's), apart from the
operator account's error counts in /var/lib/zfs-tools, and refuses a folder that is a link, open to
others or someone else's, and a file that is a link: a new file opened O_NOFOLLOW|O_EXCL, renamed over.
With mesh-controller #170 a file that asks for a setting counts as trusted
unless it says otherwise, and its settings are set at the controller's
terminal alone (hq issue 339). These two are a person's own notifier look and
folder names in their home, which nothing else trusts.
The second review of #147 found three blockers before the home server.
1. ${setting:} was not filled in a unit's name, so the distribution timers would have reached the
machine as "zfs-scrub-weekly@${setting:scrub-pool}.timer". mesh-controller #171 fills it and refuses
any value that is not a unit-name part. The module's own schedule needs no setting in a name now: the
node-engine's timer for the scheduled process zfs-mesh-scrub fires `zfs-tools scrub-run` every quarter
of an hour, and the run reads the pool, cadence, day and time from the configuration file.
2. The run does what upstream's zfs-scrub@.service does: if a scrub is running it waits for it with
zpool wait -t scrub, otherwise it starts one and waits. It never pauses or stops a scrub, and the unit
has no ExecStop. A resilver defers the scrub. The served slot is kept in /var/lib/zfs-tools.
3. Unassigning left the pool with no schedule. mesh-host #66 (hq ADR 0269, proposed) gives the
distribution timer back running and enabled. The README says not to unassign before it is merged.
Also: the run comes before the distribution timers in the declaration; the day and time are checked
against fixed sets and reach no unit file (an invalid value runs nothing); a due scrub not started an
hour after its time is a warning; the error counts outlive a restart in a small state file.
The safety review of the first version found it safe for the pool's data and asked for these fixes before
it goes on the home server:
- zfs_snapshot: no recursive. Every answer gives the dataset's free space and the pool's capacity, and
says that the snapshot holds every block deleted or overwritten until it is destroyed by hand.
- A command that runs out of time is ended with its whole process group. An act that timed out is
answered from the list, or as "outcome unknown", never as "NOT DONE".
- The watch runs as the operator account.
- The event follower reads ZFS's queue from its start, with a cursor per boot, so the pool's import and
any downtime reach the journal once. It writes "dropped N events", kills zpool before waiting for it,
and exits non-zero after five failed follows in a row.
- Past error counts are a warning, and counts that rise are unhealthy.
- The guard test catches .Run(, exec.Command(, exec.CommandContext( and process starts.
- The fixtures and the parser follow OpenZFS 2.4.4's own printers: human scan sizes under -p, no iostat
section lines, error scrubs, per-vdev rebuild resilvers, paused scrubs.
The operator also asked the module to own the scrub. A timer and a service of the module's own run
zpool scrub -w -- <pool> from the scrub-pool, scrub-cadence, scrub-day and scrub-time settings. In the
same declaration, the distribution's weekly and monthly timers for that pool are held stopped and
disabled, so exactly one schedule scrubs it. zfs_scrub shows the schedule and the next run, with any
resilver's progress. zfs_scrub_start, zfs_scrub_pause and zfs_scrub_stop act with why. The health check
warns of a late or failed scrub and of a running resilver, and is unhealthy for a resilver that finished
with errors.
2026-10-09 01:17:34 +02:00
1403 changed files with 174658 additions and 10516 deletions
basePath:=flag.String("base","","the trunk's copy of unmoved (the merge base's); seed when absent")
flag.Parse()
root:="."
ifflag.NArg()>0{
root=flag.Arg(0)
}
found,err:=Check(root)
iferr!=nil{
fmt.Fprintln(os.Stderr,err)
os.Exit(1)
}
unmoved,err:=Unmoved(listFile)
iferr!=nil{
fmt.Fprintln(os.Stderr,err)
os.Exit(1)
}
base:=seed
if*basePath!=""{
ifbase,err=Unmoved(*basePath);err!=nil{
fmt.Fprintln(os.Stderr,err)
os.Exit(1)
}
}
refused,stale,added:=Judge(found,unmoved,base)
for_,m:=rangerefused{
fmt.Printf("%s says %s, the settings form's names, and is not on %s: write config, takes-config and "+
"${config:<key>} throughout, and name the keys each mergeable file takes (novox/hq ADR 0348 §1, §11)\n",m,strings.Join(found[m],", "),listPathInCatalog)
}
for_,m:=rangestale{
fmt.Printf("%s names %s, which says no old name: take it off the list\n",listPathInCatalog,m)
}
for_,m:=rangeadded{
fmt.Printf("%s gains %s, which the trunk's copy does not hold: the list only shrinks — move the module to "+
fmt.Printf("%s names %q (%s) under ~/.ssh: only %s may, and the sweep moves anything else out (novox/hq ADR 0415 §6)\n",f.Module,f.Value,f.Key,SSHModule)
manifest(t,root,"quiet",`{"module":"quiet","description":"reads ~/.ssh/config","resources":[{"path":"${machine:account-home}/.sshrc-not","template":"# see ~/.ssh/config"},{"path":"/etc/ssh/sshd_config"}]}`)
iflen(found)!=1||found[0]!="m/a.go:4: `mesh-controller push` is a daemon's line — the operator types `nox …` (novox/hq ADR 0272) — in \"say `mesh-controller push`\""{
t.Fatalf("found %q",found)
}
}
// The list's lines read as `<daemon> <command>`, a trailing ` …` meaning an argument must follow; anything else is
forge="${MESH_FORGE:-$(printf'%s\n'"$origin"| sed -n 's#^\(https\{0,1\}://[^/]*\)/.*#\1#p')}"
owner="${MESH_FORGE_OWNER:-$(printf'%s\n'"$origin"| sed -n 's#^https\{0,1\}://[^/]*/\([^/]*\)/.*#\1#p')}"
if[ -z "$forge"]||[ -z "$owner"];then
echo"NOT CHECKED: no TypeScript toolchain at $deps, and no forge to assemble one from (origin is '$origin'; set MESH_FORGE and MESH_FORGE_OWNER, or TOOLCHAIN_MODULES)"
exit1
fi
echo"no TypeScript toolchain at $deps: assembling one from $forge/$owner/mesh-tools node-tools/ and $owner's package registry (novox/hq issue 742)"
# The catalogue's own check (novox/hq ADR 0237 as amended): the second layer of a pull request's merge
# check, `mesh/repo-check`, run by the build seat in the mesh's Go toolchain.
@@ -13,8 +14,13 @@
# it shares a rule with refused (MESH_GATE is the controller the mesh runs) — and the count of
# long-running resources without `health` held to the number in health-undeclared, which only goes down;
# 2. the Go tests of every module the change touches that has them, under the race detector when the
# toolchain has a C compiler — and a module whose dependencies cannot be fetched here, or that is
# written in TypeScript, is said as not tested, never passed silently;
# toolchain has a C compiler. A requirement of the mesh's own forge the module cache does not hold (a
# pinned SDK commit) is taken from its repository's clone beside the check (checks/deps); a module whose
# dependencies still cannot be had **fails the check**, by name and why — its tests could not run, and a
# test that could not run is a failure, never a note (novox/hq issue 385). The integration tests (`//go:build
# integration`) run here, with -tags integration, and only here. A module written in
# TypeScript is type-checked and tested by
# merge-check-typescript.sh, declared above, in the TypeScript toolchain (novox/hq issues 302, 311);
# 3. the words (novox/hq ADR 0244): no word the glossary retired for the tools, as copied in
# retired-words, in any text a module's tools can show an agent — every string in its own code that is
# not a test, and every description in its manifest. Run over every module, not only the touched ones,
@@ -22,9 +28,89 @@
# 4. the paths (novox/hq ADR 0260): no module names a path under /run, /var/lib, /usr/local/lib or /etc
# that another module's manifest places — modules talk over the bus, not through each other's files.
# Over every module; checks/paths/shared and checks/paths/known narrow it, and known only shrinks.
# 5. the agent's home (novox/hq ADR 0266): no module unpacks an archive, or writes into a file, under
# ${machine:agent-home} — the node-engine does both after checking the path, and the agent account owns
# that home and could swap a link in between.
# 6. the secrets (novox/hq issue 342): no mergeable file fills a key with a vault secret that a setting could
# replace — the key is `protected`, or left out of the file's `takes-settings`. checks/secrets/known names
# the nested ones the controller cannot protect alone yet, and only shrinks.
# 7. the list of predefined keys (novox/hq ADR 0348 clause 7): a change to predefined.json removes no key a
# module still names, judged against the trunk's copy at the merge base. Whether a removed key still holds a
# value at the mesh or a node is mesh-config's to refuse; this check reads no store. The list itself and
# every manifest against it are judged by the controller's module check in step 1.
# 8. the config form (novox/hq ADR 0348 clauses 1 and 11): no module says `settings`, `takes-settings` or
# `${setting:<key>}`, in its manifest or a template it names, or has a mergeable file naming no keys (which
# takes its whole layer), unless checks/config/unmoved lists it as not yet moved; that list only shrinks,
# judged against the trunk's copy at the merge base.
# 9. the file text (novox/hq ADR 0349 §2): no module keeps a file's text inline in its manifest (`content`)
# unless checks/templates/inline lists it as not yet moved; each names its template in files/. That list
# only shrinks, judged against the trunk's copy at the merge base.
# 10. the mail-store seat (novox/hq ADR 0378 §3, issue 800): no manifest but modules/mesh-mail claims it.
# 11. owned paths (novox/hq ADR 0422 §8, issue 1024; it holds ~/.ssh for ssh-client as ADR 0415 §6 asked):
# every directory and archive says `unknown` (move-out, module or program) and a valid `knows`; `move-out`
# only under a home; and no module declares a path inside a directory of another module.
# checks/sshdir (ADR 0415 §6) stays while main runs it; it goes under issue 1035 once main runs checks/owned.
set -eu
(cd checks/words && go test -count=1 . && go run . ../..)
# The agent's home (novox/hq ADR 0266): no archive, and no file written into, under it.
(cd checks/agenthome && go test -count=1 . && go run . ../..)
# The mail-store seat (novox/hq ADR 0378 §3): only mesh-mail may claim it.
(cd checks/mailstore && go test -count=1 . && go run . ../..)
# ~/.ssh (novox/hq ADR 0415 §6): only ssh-client may name a path under it. Main's merge check runs this line, so
# checks/sshdir stays until main runs checks/owned; it goes under issue 1035.
(cd checks/sshdir && go test -count=1 . && go run . ../..)
# Owned paths (novox/hq ADR 0422 §8): unknown files named per directory; no path inside another module's directory.
(cd checks/owned && go test -count=1 . && go run . ../..)
# The secrets (novox/hq issue 342): a password is not a setting, so no setting can replace the vault's. The
# known list only shrinks: judged against the trunk's copy at the merge base, a line it does not hold is refused.
secrets_base=""
secrets_ref="origin/${MESH_CHECK_BASE:-main}"
ifmerge_base=$(git merge-base HEAD "$secrets_ref" 2>/dev/null);then
if git cat-file -e "$merge_base:checks/secrets/known" 2>/dev/null;then
secrets_base=$(mktemp)
git show "$merge_base:checks/secrets/known" >"$secrets_base"
fi
else
echo"NOT COMPARED WITH THE TRUNK: no merge base with $secrets_ref; checks/secrets/known is held to the list it was first written with"
fi
(cd checks/secrets && go test -count=1 . &&if[ -n "$secrets_base"];then go run . -base "$secrets_base" ../..;else go run . ../..;fi)
# The list of predefined keys (novox/hq ADR 0348 clause 7): no key removed that a module still names. The trunk's
# list at the merge base is what it is judged against; a trunk without the file had none.
predefined_base=""
ifmerge_base=$(git merge-base HEAD "origin/${MESH_CHECK_BASE:-main}" 2>/dev/null);then
predefined_base=$(mktemp)
if ! git show "$merge_base:predefined.json" >"$predefined_base" 2>/dev/null;then
echo'{}' >"$predefined_base"
fi
else
echo"NOT COMPARED WITH THE TRUNK: no merge base with origin/${MESH_CHECK_BASE:-main}; no removal from predefined.json can be seen"
fi
(cd checks/predefined && go test -count=1 . &&if[ -n "$predefined_base"];then go run . -base "$predefined_base" ../..;else go run . ../..;fi)
# The config form (novox/hq ADR 0348 §1, §11): a moved module says no old name, and the list of modules not yet moved
# only shrinks, judged against the trunk's copy at the merge base; a trunk without one is the list's first copy.
unmoved_base=""
ifmerge_base=$(git merge-base HEAD "origin/${MESH_CHECK_BASE:-main}" 2>/dev/null);then
if git cat-file -e "$merge_base:checks/config/unmoved" 2>/dev/null;then
unmoved_base=$(mktemp)
git show "$merge_base:checks/config/unmoved" >"$unmoved_base"
fi
else
echo"NOT COMPARED WITH THE TRUNK: no merge base with origin/${MESH_CHECK_BASE:-main}; checks/config/unmoved is held to the list it was first written with"
fi
(cd checks/config && go test -count=1 . &&if[ -n "$unmoved_base"];then go run . -base "$unmoved_base" ../..;else go run . ../..;fi)
# The file text (novox/hq ADR 0349 §2): a file's text is a template in files/, outside the modules not yet moved; the
# list of those only shrinks, judged against the trunk's copy at the merge base, a trunk without one being its first.
inline_base=""
ifmerge_base=$(git merge-base HEAD "origin/${MESH_CHECK_BASE:-main}" 2>/dev/null);then
if git cat-file -e "$merge_base:checks/templates/inline" 2>/dev/null;then
inline_base=$(mktemp)
git show "$merge_base:checks/templates/inline" >"$inline_base"
fi
else
echo"NOT COMPARED WITH THE TRUNK: no merge base with origin/${MESH_CHECK_BASE:-main}; checks/templates/inline is held to the list it was first written with"
fi
(cd checks/templates && go test -count=1 . &&if[ -n "$inline_base"];then go run . -base "$inline_base" ../..;else go run . ../..;fi)
# The lists that narrow the paths check may not grow unseen: the base branch's copies are what they are
# judged against (MESH_CHECK_BASE, main when unset), read from the clone the check runs in.
base_lists=""
@@ -70,16 +156,26 @@ if [ -n "${MESH_CHECK_MODULES:-}" ]; then
else
touched=$(printf'%s\n'"${MESH_CHECK_CHANGED:-}"| tr ',''\n'| sed -n 's#^modules/\([^/]*\)/.*#\1#p'| sort -u)
fi
# A module's dependencies: fetched, or a forge requirement taken from its clone beside the check
# (MESH_CHECK_BESIDE); its tests run against them (checks/deps, novox/hq issue 385). The helper's own test
# runs first.
(cd checks/deps && go test -count=1 .)
deps=$(mktemp -d)
trap'rm -rf "$deps"' EXIT
(cd checks/deps && go build -o "$deps/deps" .)
untestable=""
for m in $touched;do
[ -d "modules/$m"]||continue
if[ ! -f "modules/$m/go.mod"];then
[ -f "modules/$m/package.json"]&&echo"NOT TESTED HERE: modules/$m is TypeScript; the gate judges its manifest"
[ -f "modules/$m/package.json"]&&echo"modules/$m is TypeScript: merge-check-typescript.sh type-checks and tests it"
continue
fi
if ! (cd"modules/$m"&&GOPRIVATE=git.novox.be go mod download >/dev/null 2>&1);then
echo"NOT TESTED: modules/$m— its dependencies cannot be fetched by the build seat"
if ! modfile=$(GOPRIVATE=git.novox.be "$deps/deps" -beside "${MESH_CHECK_BESIDE:-}" -work "$deps""modules/$m");then
echo"FAILED: modules/$mcould not be tested — $modfile"
untestable="$untestable modules/$m"
continue
fi
[ -n "$modfile"]&&echo"modules/$m takes its forge dependencies from the clones beside the check"
echo"testing modules/$m"
unformatted=$(cd"modules/$m"&& gofmt -l .)
if[ -n "$unformatted"];then
@@ -88,6 +184,12 @@ for m in $touched; do
fi
cgo=0
[ -n "$race"]&&cgo=1
(cd"modules/$m"&&CGO_ENABLED=$cgoGOPRIVATE=git.novox.be go vet ./... &&
CGO_ENABLED=$cgoGOPRIVATE=git.novox.be go test$race -count=1 ./...)
# The integration tests (a file whose first line is `//go:build integration`: a test that starts a bus, a store or
# another process) run here and nowhere else: the unit suite stays under a second (novox/hq playbook 07).
(cd"modules/$m"&&CGO_ENABLED=$cgoGOPRIVATE=git.novox.be go vet -tags integration ${modfile:+"-modfile=$modfile"} ./... &&
CGO_ENABLED=$cgoGOPRIVATE=git.novox.be go test -tags integration ${modfile:+"-modfile=$modfile"}$race -count=1 ./...)
done
if[ -n "$untestable"];then
echo"the check fails: the tests of$untestable could not run (novox/hq issue 385)"
| `~/.icons/default/index.theme` | owned | [`config/cursor-index.theme`](config/cursor-index.theme): the cursor for programs that read neither `XCURSOR_THEME` nor the resources |
| `~/.icons/default/index.theme` | owned | [`files/cursor-index.theme`](files/cursor-index.theme): the cursor for programs that read neither `XCURSOR_THEME` nor the resources |
**`qt6ct.conf` is the mesh's now.** A change made in qt6ct's own window is replaced at the next push,
and the window's saved geometry goes with it. The theme is this module's to say. Settings will make it
"code":"! adwaita: the cursor, for X programs that take it from the resources.\nXcursor.theme: Adwaita\nXcursor.size: 24\n"
"collection":"environment",
"name":"gtk-theme",
"attributes":{
"kind":"variable",
"name":"GTK_THEME"
},
"template":"files/env-gtk-theme"
},
{
"for":"xinitrc",
"slot":"normal",
"code":"# The appearance (module adwaita): GSettings is where the portal reads dark or light, and the portal is\n# the only way it reaches Electron, Chromium, Firefox and flatpaks. Set at every session start to the\n# module's default; adwaita_appearance switches it for a session.\ngsettings set org.gnome.desktop.interface color-scheme 'prefer-dark' || true\ngsettings set org.gnome.desktop.interface gtk-theme 'Adwaita' || true\ngsettings set org.gnome.desktop.interface icon-theme 'Adwaita' || true\ngsettings set org.gnome.desktop.interface cursor-theme 'Adwaita' || true\ngsettings set org.gnome.desktop.interface cursor-size 24 || true\ngsettings set org.gnome.desktop.interface font-name 'Inter 11' || true\ngsettings set org.gnome.desktop.interface monospace-font-name 'JetBrainsMono Nerd Font 11' || true\n"
"content":"# Written by the mesh (module adwaita, novox/hq ADR 0208), for GTK 3 and GTK 4 alike. Replaced at\n# every push; adwaita_appearance switches dark and light for the running session.\n[Settings]\ngtk-theme-name=Adwaita\ngtk-icon-theme-name=Adwaita\ngtk-cursor-theme-name=Adwaita\ngtk-cursor-theme-size=24\ngtk-font-name=Inter 11\ngtk-application-prefer-dark-theme=1\n"
"content":"# Written by the mesh (module adwaita, novox/hq ADR 0208), for GTK 3 and GTK 4 alike. Replaced at\n# every push; adwaita_appearance switches dark and light for the running session.\n[Settings]\ngtk-theme-name=Adwaita\ngtk-icon-theme-name=Adwaita\ngtk-cursor-theme-name=Adwaita\ngtk-cursor-theme-size=24\ngtk-font-name=Inter 11\ngtk-application-prefer-dark-theme=1\n"
"content":"# Written by the mesh (module adwaita, novox/hq ADR 0208). Replaced at every push.\n#\n# Which portal backend answers each interface. i3 is not a desktop xdg-desktop-portal knows, so with\n# no preference it uses whichever backend happens to be installed: fine while gtk is the only one,\n# wrong the day another arrives as somebody else's dependency. Named instead. gtk also serves\n# org.freedesktop.appearance (dark or light) from GSettings, which the session's start sets.\n[preferred]\ndefault=gtk\n# Secrets for sandboxed programs come from the keyring's backend. Without this line no backend answers\n# the interface: gtk does not implement it, and gnome-keyring's names only GNOME as its desktop.\norg.freedesktop.impl.portal.Secret=gnome-keyring\n"
"content":"# Written by the mesh (module adwaita, novox/hq ADR 0208): the default cursor theme, for programs that\n# read neither XCURSOR_THEME nor the X resources.\n[Icon Theme]\nName=Default\nInherits=Adwaita\n"
Tools that say what the mesh's artifact store holds, and what the controller's records say of it
(novox/hq ADR 0251 §1–3, to-be 51). It declares no resources and changes nothing on its machine.
(novox/hq ADR 0251 §1–3, to-be 51). It changes nothing on its machine; its only resources are its state
directory and the file the mesh writes there from its setting `older-store-names`.
## Why a module beside the store
@@ -18,7 +19,7 @@ on that machine.
|---|---|---|
| `artifact_store_repositories` | r | every repository: tags and what each names, how many manifests (tagged or not), size |
| `artifact_store_usage` | r | the store's size, the largest repositories, bytes shared between repositories, bytes no manifest marks (what the nightly collector frees next) |
| `artifact_store_references` | r | what the controller's records say of each manifest, counted and sized by state and repository; each manifest listed when one repository is asked; what the records keep that the store does not hold |
| `artifact_store_references` | r | what the controller's records and the catalogue's definitions say of each manifest, counted and sized by state and repository; each manifest listed when one repository is asked; what the records keep that the store does not hold |
| `artifact_store_collect` | a | a dry run unless `dry_run` is false: what the controller would let go of, and the bytes the nightly collector would free then and now. A real run needs `why` and asks the controller's `collect` |
### How the store is read
@@ -43,6 +44,7 @@ it lists. The store's collector removes every blob no manifest marks, so these n
| state | what | removable |
|---|---|---|
| `kept` | a definition names it, or one of the five most recent builds of a module the mesh holds; `why` says which | no |
| `named-by-a-definition` | no record names it, and a definition in the catalogue pins it by digest, under the store's current name or an older one; `why` names the module (hq issue 323) | no |
| `holder-of-kept-archive` | the manifest that keeps a kept archive's blob (hq issue 253) | no |
| `eligible` | the mesh made it and keeps it for no reason | through the controller's `collect` |
| `holder-of-eligible-archive` | the manifest that keeps an eligible archive's blob | with its archive |
@@ -56,6 +58,20 @@ its own right; the controller lets an index go with the platform manifests only
counted as the index's. Two indexes naming one platform give it the more keeping state. An index no
record names leaves its platform manifests unrecorded with it.
**What the definitions pin** (hq issue 323). A definition may pin an image of the store the mesh did not
make: pushed before the mesh built it, and pinned under the store's current name or a name it had before.
No record of the controller names it, yet a machine runs it, so `artifact_store_references` reads every
definition the catalogue holds (`mesh-catalog.catalog_modules`, then `catalog_module` for each, eight at
a time and within twenty seconds) and finds every `<name>/<repository>@sha256:…` in any of its strings,
under each of the store's names: `artifact-store://`, the store's address as the controller answers it,
and the names it had before. Those are the mesh's setting `older-store-names` (space-separated, `none` by
default, set with the controller's `settings`), which the mesh writes to the module's `store-names` file: a
catalogue names no installation. When they cannot be read, the answer's first line says so. A manifest no record
names that a definition pins is `named-by-a-definition`; one the records name keeps the records' state. A
definition read at its commit is kept, so the next call asks only for the list. Definitions that could
not be read are listed under `definitions.unread`, and the answer's first line says that what they pin
reads as unrecorded until they are.
**Unrecorded copies of upstream images** are listed in `unrecorded_mirrors`: each manifest at the top of
a repository only the build's mirror writes (`upstream/<host>/<path>`, or a module's `<module>/on-<argument>`
from before hq ADR 0257) that no record names, with what it and its platforms mark alone. They are the
@@ -88,6 +104,14 @@ with slashes, a cut listing refused), manifests read with one absent or unreadab
digest never becoming a path outside the blobs, the mark (an index, a shared blob, a blob nothing marks),
every state, what the records keep that the store lacks, bytes freed counting a blob shared with a kept
manifest as kept, a dry run never confirming, a real run without `why` refused before anything is asked,
the fall-back when the references let go of cannot be carried, escalation through `sudo -n`, and that the
tools served are exactly the manifest's `tools` and its `invokes` exactly the two controller verbs.
the fall-back when the references let go of cannot be carried, a manifest a definition pins under the
store's current or an older name, a definition not read said first, escalation through `sudo -n`, and that the
tools served are exactly the manifest's `tools` and its `invokes` exactly the two controller verbs and the
catalogue's two tools.
Any command other than the two read-only scripts fails the test.
## Files it does not know
Each directory or archive this module declares says how it treats a file it did not write (ADR 0422, ADR 0416).
-`${dir:state}`: everything in it is the program's; the mesh moves nothing out (ADR 0422 §2).
"description":"The names the artifact store had before, as host names or host:port separated by spaces, or none when it had no other. artifact_store_references reads every definition under these as well as under the store's current name, so an image a definition pins under one is said to be named by a definition (novox/hq issue 323).",
"content":"# Managed by the mesh (module asus-zephyrus-g14). Replaced on every push; edit the catalogue instead.\n#\n# The discrete GPU's driver options on the ROG Zephyrus G14 (GA403, RTX 40 series, hybrid graphics).\n#\n# NVreg_DynamicPowerManagement=0x00 turns runtime D3 off. With it on, a change of power source sends\n# the driver an ACPI notification it fails to handle on this model (\"RmHandleDNotifierEvent: Failed to\n# handle ACPI D-Notifier event, status=0x62\"), and the GPU stops making progress until the machine is\n# powered off. Off costs a few idle watts in hybrid mode and keeps the machine up.\n#\n# NVreg_PreserveVideoMemoryAllocations=1 saves video memory across suspend, so what used the GPU still\n# works after waking. It needs nvidia-suspend, -hibernate and -resume to run around a sleep, which this\n# module's drop-ins on the sleep services ask for.\n#\n# A change here applies when the driver next loads: at the next boot.\noptions nvidia NVreg_PreserveVideoMemoryAllocations=1\noptions nvidia NVreg_DynamicPowerManagement=0x00\n"
"content":"# Managed by the mesh (module asus-zephyrus-g14). Replaced on every push; edit the catalogue instead.\n#\n# The ACPI video driver does not change the backlight itself on the brightness keys: on this model it\n# moves the wrong one. The keys are triggerhappy's (see /etc/triggerhappy/triggers.d/asus-g14.conf).\n# Applies when the module next loads: at the next boot.\noptions videobrightness_switch_enabled=0\n"
"content":"# Managed by the mesh (module asus-zephyrus-g14). Replaced on every push; edit the catalogue instead.\n#\n# The NVIDIA driver's own sleep actions, asked for by the sleep itself rather than enabled as\n# links: the mesh declares files and never makes links (novox/hq ADR 0012), and the host's service\n# shape must not start these units by hand, which would put the GPU to sleep with the machine awake.\n[Unit]\nWants=nvidia-suspend.service nvidia-resume.service\n"
"content":"# Managed by the mesh (module asus-zephyrus-g14). Replaced on every push; edit the catalogue instead.\n#\n# The NVIDIA driver's own sleep actions, asked for by the sleep itself rather than enabled as\n# links: the mesh declares files and never makes links (novox/hq ADR 0012), and the host's service\n# shape must not start these units by hand, which would put the GPU to sleep with the machine awake.\n[Unit]\nWants=nvidia-hibernate.service nvidia-resume.service\n"
"content":"# Managed by the mesh (module asus-zephyrus-g14). Replaced on every push; edit the catalogue instead.\n#\n# The NVIDIA driver's own sleep actions, asked for by the sleep itself rather than enabled as\n# links: the mesh declares files and never makes links (novox/hq ADR 0012), and the host's service\n# shape must not start these units by hand, which would put the GPU to sleep with the machine awake.\n[Unit]\nWants=nvidia-suspend-then-hibernate.service nvidia-resume.service\n"
"content":"# Managed by the mesh (module asus-zephyrus-g14). Replaced on every push; edit the catalogue instead.\n#\n# nvidia-powerd (Dynamic Boost) was the first error in the chain that hung this model's GPU on a change\n# of power source, and asusd starts it on mains. It runs only when the kernel command line says\n# zephyrus.nvidia-powerd — an explicit opt-in, at boot.\n[Unit]\nConditionKernelCommandLine=zephyrus.nvidia-powerd\n"
"content":"# Managed by the mesh (module asus-zephyrus-g14). Replaced on every push; edit the catalogue instead.\n#\n# The backlights are writable by the video group, so the brightness keys and the module's brightness tool\n# move the panel without root.\nACTION==\"add\", SUBSYSTEM==\"backlight\", RUN+=\"/usr/bin/chgrp video /sys/class/backlight/%k/brightness\", RUN+=\"/usr/bin/chmod g+w /sys/class/backlight/%k/brightness\"\n"
"content":"# Managed by the mesh (module asus-zephyrus-g14). Replaced on every push; edit the catalogue instead.\n#\n# On low battery the machine suspends rather than powering off, at 7 % — s2idle still draws a little,\n# so it leaves headroom. A drop-in over the package's own UPower.conf, which stays the package's.\n[UPower]\nUsePercentageForPolicy=true\nPercentageLow=15.0\nPercentageCritical=10.0\nPercentageAction=7.0\nCriticalPowerAction=Suspend\nAllowRiskyCriticalPowerAction=true\n"
"content":"# Managed by the mesh (module asus-zephyrus-g14). Replaced on every push; edit the catalogue instead.\n#\n# The touchpad's settings, applied by X every time the device appears — at login and after every\n# resume, when the device is initialised again. This replaces the predecessor's sleep hook, which ran\n# xinput after a resume as a named person on a guessed display.\nSection \"InputClass\"\n Identifier \"asus-zephyrus-g14touchpad\"\n MatchIsTouchpad \"on\"\n Option \"Tapping\" \"on\"\n Option \"NaturalScrolling\" \"true\"\n Option \"AccelSpeed\" \"0.15\"\nEndSection\n"
"content":"# The ROG Zephyrus G14's vendor keys, which reach no X client: media (the M-keys), panel brightness,\n# and the touchpad key. Each runs this module's own script, as the operator's account.\nKEY_PROG1\t1\t/usr/local/lib/asus-zephyrus-g14/bin/zephyrus-session /usr/local/lib/asus-zephyrus-g14/bin/zephyrus-media play-pause\nKEY_PROG3\t1\t/usr/local/lib/asus-zephyrus-g14/bin/zephyrus-session /usr/local/lib/asus-zephyrus-g14/bin/zephyrus-media previous\nKEY_PROG4\t1\t/usr/local/lib/asus-zephyrus-g14/bin/zephyrus-session /usr/local/lib/asus-zephyrus-g14/bin/zephyrus-media next\nKEY_BRIGHTNESSDOWN\t1\t/usr/local/lib/asus-zephyrus-g14/bin/zephyrus-backlight -\nKEY_BRIGHTNESSDOWN\t2\t/usr/local/lib/asus-zephyrus-g14/bin/zephyrus-backlight -\nKEY_BRIGHTNESSUP\t1\t/usr/local/lib/asus-zephyrus-g14/bin/zephyrus-backlight +\nKEY_BRIGHTNESSUP\t2\t/usr/local/lib/asus-zephyrus-g14/bin/zephyrus-backlight +\nKEY_F21\t1\t/usr/local/lib/asus-zephyrus-g14/bin/zephyrus-touchpad reset\n"
},
{
"seat":"node-display-session",
"kind":"config",
"content":"# The laptop's own lines in i3 (module asus-zephyrus-g14, novox/hq ADR 0208, ADR 0210). Owned by the\n# mesh: replaced at every push. Once the controller places contributions to node-display-session\n# (ADR 0210), these become the module's contribution instead of a file in i3's directory.\n#\n# The keys the firmware turns into ordinary key presses. The vendor keys that reach no X client are\n# triggerhappy's (/etc/triggerhappy/triggers.d/asus-g14.conf): M4 and Fn+F4/F5 for media, Fn+F7/F8 for\n# the panel, Fn+F10 for the touchpad. The keyboard backlight (Fn+F2/F3) is the firmware's and asusd's.\n\n# Fn+F6, the screenshot key: the firmware sends Super+Shift+S. Released before it runs, because the\n# screenshot grabs the pointer to select a region, which fails while the key is still held.\nbindsym --release $mod+Shift+s exec --no-startup-id $XDG_CONFIG_HOME/i3/scripts/screenshot.sh\n\n# Fn+F9, the display key: the firmware sends Super+P. Odd workspaces to the panel, even ones to the\n# external output.\nbindsym $mod+p exec --no-startup-id /usr/local/lib/asus-zephyrus-g14/bin/zephyrus-display order\n\n# The keyboard backlight's level, shown when it changes.\nexec --no-startup-id /usr/local/lib/asus-zephyrus-g14/bin/zephyrus-kbd-notify\n"
},
{
"seat":"node-display-session",
"kind":"config",
"content":"# The laptop's own lines in i3 (module asus-zephyrus-g14, novox/hq ADR 0208, ADR 0210). Owned by the\n# mesh: replaced at every push. Once the controller places contributions to node-display-session\n# (ADR 0210), these become the module's contribution instead of a file in i3's directory.\n#\n# The model's panel and touchpad.\n\n# The internal panel is the primary output, where the bars' tray goes. Which monitors are on and where\n# is the display server's (autorandr, run at every session start).\nexec --no-startup-id /usr/local/lib/asus-zephyrus-g14/bin/zephyrus-display primary\n\n# The touchpad's settings again, by hand. X applies them itself whenever the device appears.\nbindsym $mod+Shift+x exec --no-startup-id /usr/local/lib/asus-zephyrus-g14/bin/zephyrus-touchpad reset\n"
}
],
"shell":[
{
"for":"after-wake",
"slot":"normal",
"code":"# The touchpad's settings again after waking, in the operator's session: X applies the module's\n# input class when the device appears, and this covers a wake that does not initialise it again.\n# Runs as root from the power module; the reset itself runs as the session's owner.\nsleep 2\nowner=$(ps -o user= -C i3 | head -n 1)\n[ -n \"$owner\" ] && runuser -u \"$owner\" -- /usr/local/lib/asus-zephyrus-g14/bin/zephyrus-touchpad reset\ntrue\n"
"description":"The Host header the module's tools send to Baserow, a host name; the file's default is the name the module's route was given.",
"trusted":true,
"default":"${bound:route:name}"
},
"email":{
"type":"string",
"description":"The email address of the Baserow account the module's tools sign in as; unset, the code falls back to its environment.",
"trusted":true,
"default":""
},
"url":{
"type":"string",
"description":"The address the module's tools reach Baserow at, as a URL; unset, the code falls back to its environment, else this machine's Baserow on port 80.",
| Owl's xpi (`owl`, pinned by the vendor's sha256) | `/opt/betterbird/owl` | owned, whole |
| the module's defaults (`install/defaults-pref/mesh-defaults.js`: the preferences kept at Betterbird's own default, no value of the operator's) and the spelling dictionaries' default (`vendor-prefs.js`) | `/opt/betterbird/defaults/pref` | owned: its two files |
| the tools' program, for the start's preparation | `/usr/local/lib/mesh-betterbird` | owned, whole |
| the start | `/usr/local/bin/betterbird` | owned, whole |
| the launcher entry | `/usr/local/share/applications/eu.betterbird.Betterbird.desktop` | owned, whole |
| the module's directory, with the settings as the node-engine writes them (`prefs.json`, `accounts.json`, `configuration.json`, `tags.json`), the records of what was adopted or created, and the filter files it replaced (`kept/`) | `~/.config/betterbird/` | owned, whole |
| the profile's `user.js`, placed by the tools | the profile | the module's, or never touched |
| the lists in the profile's `prefs.js` (accounts, outgoing servers, an adopted account's identities, the calendars' order), and the preferences of what the module created and the settings no longer hold | the profile | written into: the module's keys only, only while Betterbird is closed |
| each declared account's `msgFilterRules.dat` (setting `filters`) | the account's directory in the profile | owned, whole, only while Betterbird is closed |
| the profile's own values of the preferences the module manages, in `prefs.js` | the profile | cleared only by `betterbird_config_apply` with apply, Betterbird quit |
| the profile: mail, address books, calendars, logins | `~/.thunderbird/` | found (ADR 0182) |
**The community package `betterbird-bin` is declared absent** (ADR 0180), before anything is placed in
`/opt/betterbird`: the node-engine removes it with `pacman -R`, then unpacks the release where the package
had it. Betterbird chooses its profile by its installation's path, so the same profile opens after the
switch. A package left in place (its removal failed) holds the path, and the release is refused there by
the node-engine rather than mixed in. The module creates no link: the start, a launcher entry naming the
icon in the installation, and the dictionaries as a default preference replace the package's links.
**Every new build waits for the operator** (`upgrade: record`): the first push to a node removes the
package and replaces what a running Betterbird has open, so the operator quits Betterbird, then pushes.
An upgrade is a change to the pins, reviewed and delivered like any other (ADR 0316 §5).
**The pins.** The release is `betterbird-153.4.0esr-bb10.en-US.linux-x86_64.tar.xz` (81,620,684 bytes) and
Owl is 1.5.6.1 (430,201 bytes), each pinned by sha256 in `module.json`. The build seat fetches each once,
refuses it when the digest differs, keeps it under its sha256, and packs it into the artifact store; a node
receives the mesh's archive by digest and never asks the vendor. **Owl is never in this repository**: its
licence forbids copying it to third parties, and its vendor serves only its newest version, at one
address. Keeping the pinned copy in the mesh's private artifact store for the operator's own machines was
the operator's call (2026-10-12, ADR 0316 §6).
**Owl's range.** Owl declares which Betterbird versions it supports (`strict_min_version` to
`strict_max_version`: 115 to 156 for 1.5.6.1). `owlSupports` in `manifest_test.go` holds the range read
from each pinned xpi, and the module's tests refuse a Betterbird pin outside it: moving Owl's pin means
reading the new range there (ADR 0316 §8).
**The policy** turns Betterbird's own updater off (`DisableAppUpdate`, which Betterbird's update check
honours) and installs Owl into every profile from `file:///opt/betterbird/owl/owl.xpi` as
`force_installed`, with `updates_disabled`. A profile that already runs another Owl keeps it: the policy
installs the pinned one where Owl is missing. A newer Owl in the profile (g14 runs 1.5.7) is said as a
fact by `betterbird_status`; Owl installed by policy cannot be removed in Betterbird, so the remedy is
moving the pin. An older one is a finding of `betterbird_check`, with the same remedy.
**Betterbird never asks to be the default client.** The policy locks `mail.shell.checkDefaultClient` off
(`Preferences`, `Status: locked`): Betterbird's own default (`all-thunderbird.js`) is to ask at every start,
with a window to make it the default for mail, news, feeds and calendar files, while that preference is true
(`messenger.js`). Betterbird 153 has no policy of its own for it (its `policies-schema.json`), and its
`Preferences` policy accepts the `mail.` prefix. A lock, not a default: the window's own "always check" box
writes the preference into the profile, which wins over a default; a lock wins over the profile and greys the
box out in the settings. `betterbird_check` says when Betterbird would ask (novox/hq issue 345). **Owl's licence stays the person's**: the mesh holds no licence, key or payment (ADR 0316 §9).
**No start of its own.** Betterbird is started by the operator from the launcher. The start
(`/usr/local/bin/betterbird`) first runs `betterbird-tools prepare`, which, with no Betterbird of the
account running, places the user.js and writes what Betterbird rewrites itself (the lists, the removals, the
filter files); then it
starts Betterbird either way. `betterbird_restart` starts it as the transient user unit
`mesh-betterbird`, so its output then goes to the journal.
## The preferences: locks, the setting prefs, and Betterbird's own defaults
Betterbird's preferences come in three layers (novox/hq issue 345, issue 586). **The operator's own choices
live in the module's configuration, never in the catalogue**: the module ships only its policy's locks and the
names it keeps at Betterbird's own default.
| layer | where | who changes it | holds |
|---|---|---|---|
| **locked** | `/opt/betterbird/distribution/policies.json` (`install/distribution/`) | only a reviewed catalogue change | what must never differ on a node; Betterbird's settings show it greyed out |
| **declared** | the setting `prefs`, mesh-wide with a node's layer over it, written to `~/.config/betterbird/prefs.json` and placed in the profile's `user.js` | the operator, through the settings layers (trusted: the terminal or the operator's warrant) | the operator's setup, and a node's exceptions, such as a font size for a larger screen |
| **Betterbird's default** | Betterbird's own defaults; `mesh-defaults.js` names those the module manages there | a catalogue change | the preferences kept at Betterbird's default |
A locked value wins over everything; a declared one (in `user.js`) wins over the profile's `prefs.js` at every
start; `prefs.js` wins over a default.
**The setting `prefs` is one entry per preference** (issue 586): `{"prefs": {"mail.uifontsize": 14, …}}`, the
value true, false, a whole number or a string. The node-engine merges the layers key by key, so a node's layer
overrides one preference, or takes one out with `null`. The operator's setup approved on 2026-10-12 (compact
density, 14-point text, the vertical view with multi-line cards and threads, no new-mail pop-up, sound or tray
icon, the calendar's day from 09:00 to 20:00, reminders missed while asleep shown, chat off, remote content
blocked, Do Not Track; 21 preferences) moved from the catalogue's defaults file into the mesh-wide layer of
`prefs`, set on the operator's Approve.
**The order that leaves no node on Betterbird's own defaults** (issue 586): merge this build (the `record`
upgrade policy holds it back from the nodes); propose the mesh-wide layer through the controller's `settings`
verb with `propose`, carrying the current `accounts` verbatim and `prefs` with the 21 values; the operator
approves; only then push the desktop nodes. Caveat: between that approval and a node's push, any other send to a
desktop node writes the map's text into the old build's user.js, and `betterbird_check` there says the user.js is
refused until the push.
**Locked** (the policy): Betterbird's updater off (`DisableAppUpdate`); telemetry off (`DisableTelemetry`);
add-ons change only through the mesh's pins (`ExtensionUpdate` false); and three preferences locked off:
`mail.shell.checkDefaultClient` (the ask to be the default client), `mail.inappnotifications.enabled` (the
vendor's in-app notices), and `mailnews.start_page.enabled` (the start page, which otherwise loads from the
vendor at every start).
**Kept at Betterbird's default** (`// keeps Betterbird's default: <name>` in `mesh-defaults.js`): the theme,
system colours, always-underlined links, manual spam marking, and the composer's attachment pane. A value the
profile saved for one of them is drift.
**Drift.**`betterbird_config` shows each managed preference (locked, declared, kept) through the layers, and
flags **drift**: a declared value the profile's `user.js` does not hold yet (remedy: `betterbird_user_js` with
apply, then `betterbird_restart`), or a value the profile saved where the module keeps Betterbird's default.
`betterbird_check` reports it.
-`betterbird_config_apply` with `apply` takes a kept preference's saved value out of `prefs.js`, so Betterbird's
default holds at the next start. It acts only while Betterbird is quit (no main process, and no `lock` in the
profile), and says what it cleared.
- **The theme is the exception.** Betterbird keeps the active theme in its add-on database (`extensions.json`),
and `extensions.activeThemeID` only mirrors it. Another theme is drift whose remedy is to enable the default
theme in the add-on manager. `betterbird_config_apply` never clears it, and `keep` refuses it.
-`betterbird_config` with `keep` proposes this node's layer of `prefs` holding the profile's values of the named
preferences beside what the layer holds, through the controller's settings verb, for the operator's Approve.
**One owner per value.**`prefs` refuses a preference whose name can hold a credential, an account's
(`mail.account.`, `mail.identity.`, `mail.server.`, `mail.smtpserver.` and the account lists, ADR 0316 §11), and
those the settings below own: `calendar.registry.`, `calendar.list.`, `ldap_2.servers.` and `mailnews.tags.`
(ADR 0329 §6).
## The user.js
Betterbird reads `user.js` in its profile when it starts, and a preference set there wins over the one it saved
in `prefs.js`. The profile's directory has a random name per machine (`<random>.default-default`), which a
manifest cannot name. So the tools compose the user.js from the settings — `prefs`, then the declared accounts,
then the declared identities, calendars, address books and tags under the keys each resolved to — and place it
in the profile: at the start from the launcher, `betterbird_user_js` with `apply`, or `betterbird_restart`.
**The profile is found** as the one the running Betterbird holds open; with none running, as the
installation's default in `profiles.ini` (`[Install…] Default`), else the profile marked `Default=1`.
**A user.js the module did not write is never overwritten.** The module's begins with the line
`// Betterbird's user.js (module betterbird`; any other is the operator's, and `apply` refuses with what to do.
## The accounts: declared without credentials (ADR 0316 §10–§16, ADR 0317)
The setting `accounts` reaches the machine as `~/.config/betterbird/accounts.json`, a trusted mergeable
file: a map of accounts by name, each with `address`, `from-name` and `label` (optional), `incoming` (`kind`
imap or pop3, `host`, `port`, `security` tls, starttls or none, `sign-in` password, encrypted-password,
oauth, kerberos, ntlm or tls-certificate, `user`) and `outgoing` (the same, without a kind; its sign-in may
be none), and, only in a node's own layer, `key`. **Never a password or a token**: the whole file is
refused when any key in it is shaped like one, and an account signs in at its first use, in Betterbird's
own window, where Betterbird keeps the result in its login store. The work account through Owl is never
declared: the person creates it in Owl's setup (ADR 0316 §13).
At every apply (the start from the launcher, `betterbird_user_js` with apply, `betterbird_restart`), each
declared account, in its place in the setting, is (ADR 0317 §1):
1.**kept by its recorded keys**, when the module recorded it on this node for this profile;
2.**its creation**, when it is under the module's own keys (`account_mesh_<name>`, `id_mesh_<name>`,
`server_mesh_<name>`, `smtp_mesh_<name>`);
3.**adopted by the key a node's layer names** (`key`), when that account is in the list, its default
identity has the address and it is not Owl's;
4.**adopted by its address**, when exactly one account's default identity has it: its keys (account,
default identity, incoming server, and the outgoing server that identity names) become the module's, and
the account lists are left as they are;
5.**created** under the module's own keys when no account has it: its keys are added at the end of
`mail.accountmanager.accounts` and `mail.smtpservers` in `prefs.js`, every other key kept in its place,
only while Betterbird is not running.
**Refused, for that account alone and with the remedy** (§7): two accounts with the address (remove the
duplicate in Betterbird, or name the `key`), Owl's account at the address, a named key that does not
qualify, an account of another kind than declared (never converted), an identity that names no outgoing
server, and an outgoing server that another identity, neither adopted nor the module's, also uses
(Betterbird lets identities share one, and the declared values would change that other account's sending;
ADR 0317 §2 leaves the identity's server and the lists as they are, so the module refuses rather than give
it a server of its own). A recorded adoption follows the identity's current outgoing server at every apply,
under the same rule. The other accounts go ahead; the start and `betterbird_check` say it.
`betterbird_user_js` answers an account's preferences by name with their values withheld.
The declared fields are written in the user.js under the account's keys, so they win at every start; only
the fields the setting names, never the account's directory, folders, filters, caches, mail or kept
sign-in, so an adopted account keeps its sign-in (§5). The apply names, by preference name only, each one
whose value it changed (§4). What was adopted or created is recorded in
`~/.config/betterbird/accounts-record.json`. **Leaving the setting** releases an adopted account (its
lines and its record go, the account stays the person's) and takes a created account's keys out of the
lists (its mail stays in the profile).
**The value is derived, not retyped** (§9): `betterbird_accounts` with `propose` builds `accounts` from the
profile (every account but Owl's, no credential) and proposes it in the mesh-wide layer as it stands, through
the controller's settings verb; the operator approves on a channel that proves who answers. Its answer shows
the value it proposed (ADR 0329 §10).
**A node with no profile yet.** The preparation needs the profile Betterbird makes at its first start: on a
node set up from nothing, the declared accounts are written at the second start from the launcher (or at
`betterbird_restart`).
## Identities, calendars, address books, filters and tags (ADR 0329)
Five more settings, mesh-wide with a node's layer over each (an entry set to `null` in a node's layer is not on
that node). `identities`, `calendars`, `address-books` and `filters` reach the machine as
`~/.config/betterbird/configuration.json`, trusted as `accounts` is; `tags` as `tags.json`, trusted too since issue 751 (the operator's decision), so a change to it is the operator's to approve.
**No value holds a password, a token or a cookie**: a key shaped like one is refused, and so is an address whose
user part carries a password (§2). A setting holds no line break, so a signature is a list of its lines.
| setting | an entry | matched in the profile by |
|---|---|---|
| `identities` | by a name: `account` (its name in `accounts`), and for another identity than the account's default `address` and `name`; `reply-to`, `organization`, `signature` (lines), `signature-html`, `signature-above-quote`; `key` in a node's layer | the account's default (no `address`), or an identity of that account with the address |
| `calendars` | by a name: `name`, `url` (CalDAV), `color`, `read-only`, `reminders`, `user`, `invitations` (an identities entry); `key` in a node's layer | its address |
| `address-books` | by a name: `name`, `url` (CardDAV), `user`; `key` in a node's layer | its address |
| `tags` | by its key: `name`, `color`, `order` | its key |
| `filters` | by an account's name: the ordered list, each with `name`, `enabled`, `when` (new-mail, manual, after-junk, sending, archiving, periodic), `match` (all, any, every-message), `conditions` (`field`, `op`, `value`, in Betterbird's words) and `actions` (`action` in Betterbird's words; `folder` as a path within the account, or `value`) | owned whole per account |
**Adopted as accounts are** (§7): an entry recorded on this node is kept by its key; one under the module's own
key (`ident_mesh_<name>`, `mesh_<name>` for a calendar or an address book) is its creation; a `key` in the
node's layer is adopted when the profile holds it and it matches; else one match is adopted, none is created
under the module's key, and several are refused with the remedy. The declared fields go into the user.js under
that key. An entry leaving the setting releases what was adopted (its lines go) and removes the preferences of
what was created, never a calendar's events, an address book's contacts or a cache. An entry no setting declares
stays the person's. Owl's accounts and calendars, an ICS calendar (its address is often the secret), local
calendars and address books are never declared (§13).
**Filters are owned whole per declared account** (§8): the module writes that account's `msgFilterRules.dat`
from the list, each folder's path turned into its address on this node, so a filter added in Betterbird is gone
at the next apply (drift shows it first). Before its first write it keeps the replaced file in
`~/.config/betterbird/kept/<profile>/`; when the account leaves `filters`, the module stops writing and the file
stays the person's; `betterbird_filters` says where the kept copy is. A filter a setting cannot hold as it is (a
folder of another account, an add-on's action, a news filter) is named and not derived.
**Written only while Betterbird is closed** (§9): the lists in `prefs.js` and the filter files, because
Betterbird rewrites them as it runs and quits; the user.js is placed as before. `betterbird_check` counts per
kind the changes that wait.
**Derived, then proposed one layer at a time** (§10): `betterbird_derive` asks every desktop node holding
Betterbird for what its profile holds. An entry on every such node goes into the mesh-wide layer with the value of
the node the derive was called on, and the answer lists each field another node holds differently (which then
shows as drift there); one on some nodes only goes into those nodes' own layers. An account whose filter file
holds no filter is not put in `filters`. With `propose`, the next layer (the mesh-wide one, then each node's) is
proposed with the layer as it stands beside the derived kinds; never while a proposal of the module's waits, and
never while a desktop node does not answer. `prefs` on the call carries values of the setting `prefs` into the
mesh-wide layer. A `key` is never proposed for the mesh-wide layer.
## Tools
They are served by the node's runtime as the operator account (ADR 0175). **They show the operator's own
configuration with its values, and never a credential** (ADR 0329 §1, §2): every answer passes the module's
credential check, which withholds a field whose name says password, token, secret or cookie and the password
part of any address. No answer carries mail, a calendar entry or a contact.
| tool | does |
|---|---|
| `betterbird_status` (r) | whether it runs (pid, since, scope, children); what is installed and from where, its version beside the newest published build; the policy; Owl's version against the pin and whether it is active; where its output goes; the profile and how it was found; the user.js (declared, placed, in effect, with every declared entry as it resolved); the declared accounts by count; the add-ons; the preferences that decide how its windows and notifications look |
| `betterbird_prefs` (r) | named preferences or prefixes: value and source, and the default with the file that gives it. Withheld without its value: a name that can hold a credential, one that names an account, server, calendar or person, and every `extensions.*` but the theme's id. Values masked (the home as `~`, addresses and hosts) |
| `betterbird_config` (r; a with `keep`) | each managed preference: its lock, the declared value, Betterbird's default where the module keeps that, the profile's saved value, the value in effect, and drift; with `keep`, this node's layer of `prefs` proposed with the profile's values |
| `betterbird_config_apply` (r; a with `apply`) | a kept preference's drifting values listed; with `apply`, while Betterbird is quit, taken out of `prefs.js` |
| `betterbird_user_js` (r; a with `apply`) | the user.js composed from the settings against the profile's; `apply` places it, writes what Betterbird rewrites while it is closed, and counts what waits |
| `betterbird_log` (r) | the last lines of the journal of `mesh-betterbird` (or of the reminder test), addresses and credentials masked |
| `betterbird_restart` (a) | ends Betterbird (asked to quit, forced after 8 s), places the user.js and writes what waited, starts it as `mesh-betterbird`. **Refused while a composer window is open**, unless `discard_unsent` is true, which only the operator's yes justifies |
| `betterbird_check` (r) | installed from the mesh's release; the policy and the module's defaults placed; no managed preference drifting; no ask to be the default mail client; Owl at its pin and active; one running; profile found; the user.js valid, placed, in effect; the declared accounts in the list, none refused, each signed in once; per kind, the declared entries drifting or refused and the changes waiting for Betterbird to close; its output read |
| `betterbird_accounts` (r; a with `propose`) | the profile's accounts with their values and the declared account each is, Owl's by name and address and marked; with `propose`, `accounts` built from them and proposed in the mesh-wide layer as it stands |
| `betterbird_identities`, `betterbird_calendars`, `betterbird_address_books`, `betterbird_filters`, `betterbird_tags` (r) | the profile's entries with their keys and values beside the declared ones and how each resolved (adopted, created or refused with why); an ICS calendar by its name only, Owl's marked; `betterbird_filters` also where the kept filter files are |
| `betterbird_derive` (r; a with `propose`) | the five kinds derived from the desktop nodes' profiles, per layer, with the fields another node holds differently and what was left out; with `propose`, the next layer proposed; `here` answers this node's profile, as another node asks it |
| `betterbird_drift` (r) | per kind and declared entry: holds, differs (each field with both values), lacks, or refused; and the profile's entries nothing declares |
| `betterbird_reminder_test` (a) | a reminder on demand without the operator's calendar (below) |
### A reminder on demand
**It opens windows on the operator's screen**, and its description says so first, so a caller tells the
operator before calling it. On 2026-10-12 at 12:05 a test Betterbird with an empty profile carrying the
operator's theme looked like the operator's configuration lost; so the test is **visibly a test**: its
profile carries a `userChrome.css` that frames every window of the test Betterbird in red and runs a red
band reading MESH TEST across its top, and its event is called "MESH TEST reminder (a throw-away
Betterbird of the mesh, not your calendar)". The reminder window itself is left unstyled, because how
Betterbird draws it is what the test looks at; its title is the event's. A window title cannot be set by a
preference, so the marking is the band, the frame and the event's name.
`betterbird_reminder_test` starts a **second** Betterbird with `--no-remote` in a throw-away profile
under `/run/user/<uid>/mesh-betterbird-reminder-test/`, as the user unit
`mesh-betterbird-reminder-test` (ended after 10 minutes at most). The profile holds the module's
user.js (the setting `prefs`; never a declared account) and one calendar: a local `.ics` file with one
event, "MESH TEST reminder", whose reminder is
due `after` seconds from the call (default 40). `carry` copies named preferences of the operator's
profile into it (a theme, a colour switch), under the same limits as `betterbird_prefs`. The test's own
preferences come last in its user.js, so nothing declared or carried overrides them; one of them turns
Betterbird's update check off (`betterbird.update.checkIntervalDays` 0), because a fresh profile checks
at its start and, when a newer build is published, opens the About window on the operator's screen
(seen 2026-10-12). Then:
-`xorg_windows` with class `Calendar` answers the reminder window: role `AlarmWindow`, type `Dialog`,
its depth (32: a visual with alpha, which the compositor blends), its map state;
-`xorg_screenshot` with that window shows it as the screen shows it.
The unit removes its profile when it ends, however it ends (`ExecStopPost`): at its 10 minutes, on
`stop`, or when the test window is closed. The test Betterbird is never counted as the operator's: its
command line names its own profile, and `status`, `check`, `restart` and the user.js leave it out.
The running Betterbird, its profile and its calendars are not touched: `--no-remote` keeps the two
apart. **`profiles.ini` and `installs.ini` are not rewritten either**, measured on the workstation
(2026-10-09): their content and modification time (June 2023) were the same before a `--profile` start,
while it ran, and after it ended.
## The reminder window, as measured (2026-10-09, workstation, i3 and picom with glx)
| | |
|---|---|
| X class / instance | `eu.betterbird.Betterbird` / `Calendar` |
| role, type | `AlarmWindow`, `_NET_WM_WINDOW_TYPE_DIALOG` (i3 floats it) |
| depth | 32 (ARGB visual), like every Betterbird window there |
| size | 600×300, kept in the profile's `xulstore.json` |
A window of depth 32 that its program has not drawn is transparent under a compositor, not black:
that is why the symptom looks the way it does. Why Betterbird did not draw it is hq issue 345's
question.
## What changes when it is pushed
The module is assigned to shanks today (look-only, from issue 345's first part); g14 runs Betterbird
outside the mesh. With this version, a push to a node (each waits for the operator: `upgrade: record`):
| | on a node with the community package (g14, shanks) |
|---|---|
| the package | `betterbird-bin` removed (quit Betterbird first) |
| `/opt/betterbird` | the mesh's 153.4.0esr-bb10, in the same place, so the same profile opens |
| the policy, Owl | placed; Owl installed into the profile by Betterbird at its next start where it is missing |
| `/usr/local` | the start, the launcher entry, the tools' program |
| `~/.config/betterbird/` | `prefs.json`, `accounts.json`, `configuration.json`, `tags.json` (each empty until its setting is approved) |
| the profile | untouched until the next start from the launcher, `betterbird_user_js` with apply, or `betterbird_restart` |
| the configuration | the policy's locks at Betterbird's next start; the settings in the user.js from the next start. Pushed in the order above, the operator's 21 preferences are in `prefs` before the build arrives; pushed before that layer is approved, Betterbird runs with its own defaults for them until it is |
## Relies on
- A display server on the same machine (`x11-display`, ADR 0208 §3).
- **`xorg`** for `xorg_windows` and `xorg_screenshot`, which look at Betterbird's windows. This module
carries no window code of its own.
-`systemd-run` and `journalctl` of the account's user manager.
- The build seat's archive pinned by URL and sha256 (mesh-controller, ADR 0316 §2): a controller that does
not know it refuses this manifest.
- The controller's `settings` verb, for every proposal and to read a layer as it stands; its `nodes` and `node`
verbs, for which machines hold the module; and the module's own `betterbird_derive` on those machines
(`invokes`).
## Files it does not know
Each directory or archive this module declares says how it treats a file it did not write (ADR 0422, ADR 0416).
-`~/.config/betterbird`: holds only what this module declares and its own tools' records: `accounts-record.json` (`cmd/betterbird-tools/adoption.go:36`), `configuration-record.json` (`owned.go:33`) and the `kept/` folder of replaced filter files (`filters.go:261`); anything else is moved out by the node-engine to `~/.local/state/mesh-moved/betterbird/` with a list, never deleted (ADR 0422 §4).
-`/opt/betterbird`: everything in it is the program's; the mesh moves nothing out (ADR 0422 §2).
-`/opt/betterbird/owl`: everything in it is the program's; the mesh moves nothing out (ADR 0422 §2).
-`/opt/betterbird/distribution`: everything in it is the program's; the mesh moves nothing out (ADR 0422 §2).
-`/opt/betterbird/defaults/pref`: everything in it is the program's; the mesh moves nothing out (ADR 0422 §2).
-`/usr/local/lib/mesh-betterbird`: everything in it is the program's; the mesh moves nothing out (ADR 0422 §2).
returnjson.RawMessage(`{"ok": true, "output": "{\n \"tags\": {\"$label1\": {\"name\": \"Important\"}}\n}\n approved by the operator\nbetterbird on the whole mesh, every value"}`),nil
}
returnjson.RawMessage(`{"ok": true, "output": "proposed p-17 for betterbird on the whole mesh: accounts (other@mail.example.org); fingerprint ab12"}`),nil
returnjson.RawMessage(`{"ok": true, "output": "{\"prefs\": {\"calendar.alarms.playsound\": false}, \"identities\": {\"mail\": {\"account\": \"mail\"}}}\n set at the terminal\nbetterbird on shanks, every value"}`),nil
t.Errorf("%s's copy of %s differs from this one: change every copy together",module,f)
}
}
compared++
}
ifcompared==0{
t.Log("no sibling copies beside this module")
}
}
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.