The store is found by the seats' holder and its listen registry; list says,
per repository, each manifest's digest, whether it is an index and what it
names, its tags, its size and whether it is held, so the sweep reads the
store through the seat. E15 now asks that nothing is provided or served.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
messenger's refusal took its declaration of operator-channel out of the judged catalogue, so claude-code's use of it
was refused too. distribution's two remaining refusals (its program builds beside the artifact store it provides,
and its only health check is its own tool) are a design question left to the lead.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
Saying removed for a version the store still holds told the builder something false. remove now answers
kept with the reason shared-digest and the other tags naming its digest, removed false, and says the
seat's event kept with them (mesh-contracts' shared registry part); the builder raises no retry.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
registry_incoming, npm_incoming and go_incoming dropped a transfer by hand with only a log line. Each
drop now records its hand act through the seat mesh-hand-acts before anything goes, as nats_requests
does; an act not recorded leaves the transfer open. distribution's manifest invokes the seat.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
A transfer whose idle key ended while the bundle was down was never dropped. Both hand mesh-sdk's
FollowIdle their open transfers, which it gives Expired once the watch is taken.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
Every module that imports mesh-sdk requires its group head; each requires the mesh-contracts seats and clients it
imports, and mesh-contracts/go through the SDK's seat list. The placeholder versions are gone.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
The store's pinned image is registry 2.8.3, whose DeleteManifest cannot delete a tag alone: by a tag it
deletes the empty digest and answers 400 DIGEST_INVALID. The fakes accepted a DELETE by tag, a fixture
agreeing with the mistake, so every merge of a reproducible build would have been a refused remove and a
retry finding live. RemoveTag now answers ErrNoTagAlone for that answer; remove then keeps the tag, whose
bytes stay under the other tag anyway, and says removed with that reason, so the builder lets it go. Any
other store failure is still refused; a digest shared with nothing is still deleted by digest. The fakes
answer as 2.8.3 does.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
Both holders carried the same idle bound and the same read of incoming's cap; the second instance of the
kind moved into mesh-sdk's Receiver (Arm, Expired, Bound, with BusArm, BusCap and FollowIdle). Each
holder now hands its idle state and its objects to those, and keeps its rows.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
The state mirrors had no read tool and no ask-again (seams review note 1): a copy a follower holds and
does not end could only wait for a restart. registry_mirrors lists each kept copy and its claim, asks one
again with its claim let go, and drops one, saying refused under the ask's id so its waiter ends.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
The holder counted against 20 GiB while the bus caps incoming at 8 GiB, so the bus refused writes before
the holder said refused naming the cap, and the 80 % finding came too late (running review F2). The cap is
read from incoming's binding (the object store's fill); IncomingCap, 8 GiB, is said only where none is read,
and the manifest's row holds it equal to max-mib. An abandoned transfer was dropped only on the next publish,
and the transfers state's own lifetime dropped its key while its parts stayed: transfers keeps for good, and
each open transfer has a key in the new state idle with its own time to live, whose end drops it (re-armed
for the rest when it took a part since). registry_incoming reports the bound cap.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
A read naming no node (an empty node_haves) closed both windows for good (security review F3, second of
a kind): the window embeds delivery.WindowHeld and steps by the shared rule, named being every node the
read named, and keeps its own guard that a window once closed stays closed. A node-engine whose
manifest-reads did not answer (a timeout included) counts as stating nothing as before, and the window's
record now says so per node (Why), so registry_old_name tells a timeout from an old node-engine. A read
that fails otherwise is an error the wake judges nothing on, no longer taken as not served.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
The lead's ENG-5 ruling: on a fresh mesh the bootstrap pushes the holders' own images to the store on its
loopback with no credential, so the front's bootstrap token, its file, the bootstrap-push window and the
program's removal of the file go. Fewer secrets; nothing to refuse once a window closes.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
Deleting the manifest by its digest removed every tag on it, and reproducible builds share a digest
between a branch's tagged version and the trunk's numbered one, so a branch's end deleted the trunk's
image (security review F1). The fake catalog now deletes by digest as the store does; a store that
cannot delete a tag alone removes nothing and the removal is refused.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
distribution answers released false for what it does not hold, as the
registry seats now define, and takes the removal reason from the
generated constant; the sweep reads the answer instead of matching
the error's text.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
The sweep asks remove now instead of deleting over the store's door, so
what its door did for an index is the holder's: a removal for retention
removes the platform manifests no tagged or held index of the repository
names, and refuses a manifest such an index names.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
mesh-build-go consumes the builder's version.numbered and publishes the held
digest under its version, naming its one part by digest only, since the packed
bytes left the node; distribution tags it at the size its store holds, so the
version's manifest names the blob truly.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
The reach came from MESH_REGISTRY_REACH, which nothing sets, so every published event
answered /<name>@<digest>. The store answers the path; the holder puts the seat's name
before it, as catalogue.InRegistrySeat reads it back (ADR 0605 §9; lead's ruling on A3n).
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
The node-engine fills ${config:} only in files, unit names, owners and groups, never in a
process's environment, so MESH_CONFIG_CALLERS was never set and design 51's defaults always held.
The tools bundle now reads `callers` through stdio.Config on each call (holder.ConfiguredCallers,
Verbs.WithCallers per call); the program, which has no runtime resolving config, asks its own
tools bundle's new registry_callers on each wake of the windows (manifest invokes). The
environment variable is gone.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
A second tools process following mirrors would copy again and say published twice. The ask is claimed by
compare-and-set before it is copied, and its outcome carries the ask's id as the event id, so a redelivery is one
message. The cap's publish-refused finding stands on incoming, not on one file, and clears with the next publish
taken.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
A large base could outlast a call's bound. The ask is kept in the state mirrors and the copy runs on that state's
watch, so a copy a restart cut is taken on again; published (or refused, with why) on container-image-registry's
events is the end the caller waits on, the place the seat's contract already gives mirror.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
The holder's own transfer logic goes: parts.Receiver takes the chunks into incoming, checks the hashes, drops an idle
transfer and tells the fill; distribution gives it the builder_versions check, its store and its seats' events, so
the forge's two package seats bind the same receiver instead of a copy.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
The verbs answered the holder's own shapes (missing and ended, a part named digest, versions as bare strings, held as
the publish's event) while the seat definitions, now generated into mesh-contracts, name lacks and held, part,
versions of artifacts, tags with digests and a hold's own body. A build module using mesh-sdk's parts helper would
have read no lacking part and sent nothing. The bundle now takes and answers the contract's types and says each
event in its body; an image's manifest is the part whose digest is the publish's, as the contract names no media
type, and a hold on the image seat is of a copied base.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
The program had no test that runs it whole. This one starts run() on a nats-server of its own, with the registry
container faked on the loopback and node_haves and manifest-reads answered on their call subjects: an away node holds
both windows open, its return closes them, each closing said once on distribution.events, a direct write is then
refused, reads stay open, the health check answers through the program's port, and a waiter on published wakes on
none of the windows' events.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
An http probe of /v2/ said only that a port answered; design 51 asks the health check to call versions per seat and
probe the read side each minute, naming the program when it is down, from outside its process. registry_health asks
versions of each seat through the program's port, writes the read-side finding while it fails and clears it after.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
The windows' logic was built but nothing woke it, so both windows stayed open for ever. The program now reads every
node through node_haves and the node-engine seat's manifest-reads at its start, on each delivered and on each
manifest-reads-changed, steps both windows, says their closing once on distribution.events and raises the seven-day
finding. A build module's have counts as publishing through the seats when it runs a build heard publishing (the
tools bundle records the fingerprint of each publish's proved caller), so a paused build agent running the same
build holds nothing open.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
Until now mirror only answered a copy the store already held, so mesh-build-image had nothing to call for an upstream
base (design 51: an upstream image is copied by mirror with its pinned digest). The copy takes an anonymous pull
token, checks every manifest's and blob's bytes against its digest, copies an index's every platform first, and
fetches only the blobs the store lacks. tags was already served.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
The tools bundle runs in the tool runner's account and cannot remove a file in
distribution's 0700 state directory, so its first answered publish closes the
window bootstrap-push in the state windows and the program removes the file
(design 51 On a fresh mesh).
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
The program serves the store's port 5000 in front of the registry container, now on the node's loopback
(127.0.0.1:5001), listed before the program so the port is free when it starts (E14). The tools bundle serves
each seat's verbs as <seat>.<verb> and registry_incoming, registry_holds, registry_old_name and
registry_direct_writes, on distribution's own account, with the caller the tool runner proved. The manifest
claims container-image-registry and oci-artifact-registry with their verbs, keeps mesh-artifact-store and
provides artifact-store until the old name closes (E15), asks its bus state in requires (incoming, transfers,
holds, windows, findings, events) and declares its health outside the program. The bootstrap's one-off
credential is discarded at the first publish answered.
Not yet: the runtime's object store helper for incoming (mesh-sdk), the windows' driver on delivered and
manifest-reads-changed, the read-side health probe calling versions per seat.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
The holder answers publish and publish-part into incoming (parts of at most 1 MiB, resumed from what is held,
hashed, only the missing layers sent, 20 GiB with its 80 % finding and a refusal at the cap, a transfer idle 24
hours dropped), asks builder_versions before it stores, and says published or refused (design 51 *Publishing*,
ADR 0605 §4, ADR 0611). versions, remove (a tagged version; untagged only for the sweep's retention, which a
hold keeps), hold and release (ADR 0617 §2: the sweep alone; the operator's session by hand; never
distribution's own account), tags and mirror. The caller list is a trusted config value checked against the
caller the tool runner proved. The store's port is the program's: reads open, writes refused but while the
direct-write window is open (never over a tag published through the seat) or with the bootstrap's one-off
credential. The old-name and direct-write windows read every node, away ones included, close once, and name a
node holding them seven days.
Claude-Session: https://claude.ai/code/session_01JG73KJNrH5EgFXMKKYSrig
mesh/delivery-group group feat/774-l3-reland rejected: composed together: the checker's verdict at the members' heads (b48c90fea31f): integration: mesh-self-ch…
mesh/check failed: integration: mesh-self-check/cmd/mesh-self-check.TestTheSelfCheckWholeRunsOnTheBus: the run answered {Run:doctor-1793351737-1 At:20…
One commit with exactly the tree of the reviewed head 35b4509c0a, on main 12c2574c9d, so two compiled programs added and removed on the branch leave no trace in history (issue 1577).
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/merge-gate pass: builds distribution, new: modules/artifact-store-tools → novox; no bus step; every machine composes with the change as it did without…
mesh/delivery covered after its failure: a later walk that contains it was delivered: plan-1792260860441586760 (novox/mesh-catalog at 5e7567e5 into main,…
A module that provides the artifact store cannot build an artifact: building publishes to the store,
so the store would be needed to create itself, and the module check refuses it. The tools become
artifact-store-tools, assigned beside the store, and read manifests from the store's files through its
container as they already read the listing, so they need no address for the store's door. Named for
the artifact store, because "store" alone is the database server (hq glossary).
mesh/merge-gate fail: builds distribution → novox; no bus step; a manifest the change touches fails the module check: modules/distribution/module.json: thi…
The store had no working tools: its TypeScript client was never built, and nothing could count what
the store holds that no record names. A Go bundle lists the store's files through its own container,
reads each manifest through its door, and sets that beside the controller's records. Collection is
asked of the controller, which decides and records; the store's tools never delete. The image.pushed
event was declared and never emitted, and nothing consumes it, so it goes.
Walks run at most daily and stop after ten minutes or two million files; datasets are read from
their counters and large items from their top level only. The database platform's tables are
dumped with pg_dumpall rather than copied as live files.
Backup lines are derived from each module's data section instead of written by hand; the holder
measures declared items, reads the array under them, and deletes a retired item only after a last
restore point; the Go providers say each held consumer's size so an empty replacement is seen.
The dry run stood until the controller held each kept archive by a manifest
(mesh-controller #53) and an apply stopped reopening the collection window
(mesh-host #23, hq issue 224). The controller's collection command now
reports 134 of 134 kept archives held. hq issue 253.
The stock collector keeps only what a manifest names, and the mesh's bundles
and archives are bare blobs no manifest names: its first run, tonight at
03:30, would delete every one, the current ones included. It now reports
what it would delete and deletes nothing, until the controller holds each
kept archive with a manifest. hq issue 253.
Withdrawn 2026-10-04 to unblock novox: while-stopped named the module-local
id and the host refuses a declaration naming a container it does not have,
whole. mesh-controller#259 fixed the namespacing and it has been live since,
so the window composes as distribution.store and the host will take it.
Unchanged from before: plain garbage-collect at 03:30 with the server held
still. The store is at 40G and nothing reclaims it until this runs.
jschoubben
mesh-tools
jochen