Install the real novox server's converted service set together on one node
behind the substrate — the whole-catalogue install this rebuild never ran.
The bed loads each committed module.json from mesh-catalog (no hand-written
manifests), rewrites image refs to the scenario registry's digests, and
remaps the co-located host-port collisions (nextcloud/invoicing/route-proxy
:80, minio/invoicing :9000, gitea/umami :3000).
Proven green: the whole set of 17 modules RESOLVES and applies (191
resources); the CORE 13 converge whole — all five providers (postgres,
redis, minio, mongodb, mssql) plus keycloak, gitea, nextcloud and invoicing
reaching their providers and staying up, plus portainer, verdaccio, registry
and route-proxy.
Reported as escalated gaps (do not gate green): fail2ban (declares
capability intrusion-prevention that no host detector provides, and an
unappliable assignment blocks whole-node resolution), umami/photos/mailu
(catalog manifests do not wire the runtime/app env the images need; photos'
server image is an alpine placeholder), and firewall (nftables.service is a
oneshot that exits, but the module declares state running so mesh-host marks
it failed).
Claude-Session: https://claude.ai/code/session_01LrgweAeERJYBg88c5cKDzF
A single-node VM bed: an ollama provider and a local-model consumer are
assigned; the resolver answers the consumer's model-access with the local node
(no licence demanded), the consumer's openai.env is templated with the served
endpoint, and a request to it reaches the running model server. Proves the
node-answer of model-access end to end (ollama on host network, keyless).
Claude-Session: https://claude.ai/code/session_01LrgweAeERJYBg88c5cKDzF
A single-node VM bed: the operator sets an API key on an openai licence, the
mesh seals it to the consumer, the host unseals and mounts it, and the consumer
writes it as OPENAI_API_KEY (env + Codex auth.json). Asserts the written key
equals the one set — the other shape ADR 0050 defines, and the ADR 0054 branch
where a static-key vendor records no usage. Green on the first run.
Claude-Session: https://claude.ai/code/session_01LrgweAeERJYBg88c5cKDzF
A VM bed: a postgres provider and the model-usage consumer on one node, the
substrate on the other. A usage event injected into the mesh is upserted into
model-usage's provisioned store, asserted at both grains, latest-per-key, and
in the clear.
Also, in build-module-runtime.sh, add migrate/index.ts and pg.d.ts to the
compiled entrypoint set so a module may carry a run-once entry and an ambient
type declaration (model-usage uses the latter for the pg driver).
The bed surfaced and drove several fixes elsewhere: a short module slug for the
S3-key identity bound (ADR 0049), host-network containers getting the mesh's
names (mesh-control), and injecting the event from a publisher rather than the
pure-consumer store (its account has no publish right by design).
Claude-Session: https://claude.ai/code/session_01LrgweAeERJYBg88c5cKDzF
The bed follows the reworked flow: the manager module seals the refresh token to the node's
PUBLIC key, the HOST unseals it and mounts the cleartext at the manager's bound path, and the
refresh reads that cleartext -- no fake node key pair is mounted any more, the host uses its
own real sealing key.
- the manager is a model-access holder deployed first, so its bound facts (carrying the node
public key) are delivered; the consumer is added only once an access token exists to seal.
- adopt reads the node public key from the bound facts; the test asserts the host mounts the
cleartext refresh token for the manager, and that it reaches nowhere on the consuming node.
- the refresh_grant assertion reads { sealed, manager_key }.
Claude-Session: https://claude.ai/code/session_01LrgweAeERJYBg88c5cKDzF
A lab bed for Phase C of model-access (ADR 0050), OAuth endpoint stubbed.
It drives the real runtime images through the whole flow: the manager
seals a refresh token at rest and opens it on the manager node alone,
mesh-control is handed only the access token and an opaque re-sealed
envelope via licence submit-refresh, and the consumer writes an
access-token-only credential. Asserts the refresh token -- original and
rotated -- is nowhere on the consuming node and only ciphertext in the
control plane's database.
build-module-runtime.sh also compiles adopt/refresh/apply/usage
entrypoints. Stubbed and flagged: the vendor endpoint, the manager node's
private key (mounted; a host capability to deliver it does not exist
today), and the submit transport (the test invokes the CLI on the
manager's output).
Claude-Session: https://claude.ai/code/session_01LrgweAeERJYBg88c5cKDzF
A lavinmq provider and an amqp-ping consumer ride laptop while the
substrate's own broker owns 5672 on anchor — the twin of two-node-db.
lavinmq is the mesh's control broker AND a user-facing capability, so a
provider must publish 5672 for its consumers and cannot share a node
with the control broker that already owns it; the split unblocks the
chain single-node.
The bed proves, layered: the run-once bootstrap computed the admin hash
and wrote the broker config before the broker started (ADR 0052); the
service and both runtimes are up and stable; each module got its scoped
broker account on the substrate broker; the provisioner created the
consumer's vhost AND user, both named for the derived login; and the
consumer connected to that vhost with the mesh-minted password and
round-tripped a message. The consumer uses ${bound:amqp:as} for user
and vhost both, and the provider's serves carries the port.
Claude-Session: https://claude.ai/code/session_01LrgweAeERJYBg88c5cKDzF
A scenario and integration test assign route-proxy (provider) and hello-web
(consumer) on one node, then assert a request to the consumer's name -- sent to
the proxy -- is forwarded to the workload and returns its answer, and that
unassigning the consumer withdraws the route so the same request stops working
(the proxy replaces its table rather than merging). Modeled on
mesh-grant-end-to-end and schedule-tick: module add, assign, one push, settled,
with no module issue (route-proxy needs no scoped account).
build-route-proxy-image.sh compiles the Go proxy from
mesh-control/examples/route-proxy into mesh-route-proxy:development for the
scenario to stock. This bed proves route-forwarding over plain HTTP;
public-ACME TLS is proven separately by certificates.test.ts against a real
ACME server.
Claude-Session: https://claude.ai/code/session_01LrgweAeERJYBg88c5cKDzF
The bed that proves the scheduled-container primitive end to end. schedtest
is the thinnest carrier of ADR 0053: one container marked
schedule: "* * * * *" that appends a timestamp to a mounted data dir each
time the host fires it -- no service, no listener, no provisioner, no
runtime, no tools, no events.
The three claims it proves, from the ADR's "How each claim is checked":
installing the schedule leaves the node current WITHOUT a run (baseline
captured right after settled, the deliberate inversion of run-once); the
container fires on its cadence (a line beyond the baseline within ~150s);
and it recurs (a second line on the next minute -- cadence, not a one-shot).
schedtest serves and consumes nothing and carries no runtime, so it is not
issued a broker account: module add -> assign -> one push is the whole
sequence, no module issue. The tick image is a bare alpine served by the
scenario's registry by digest.
Claude-Session: https://claude.ai/code/session_01LrgweAeERJYBg88c5cKDzF
The GREEN multi-node regression bed that proves the DB-consumer gate: substrate/control on
one node, postgres+redis providers and baserow+letta consumers on another, each consumer
getting its own credential and its own mesh-named database across the overlay. Requires the
mesh-control provider-seal-key fix and the mesh-catalog db-name fix.
Includes a general lab capability: a machine 'disk' field sizing the VM root disk (a broad
install exhausts the pool default and the host fails mid-apply with 'no space left on
device'). The bed sets 60GiB.
Claude-Session: https://claude.ai/code/session_01LrgweAeERJYBg88c5cKDzF
Sixth green regression bed. Same shape as tools-gitlab: a runtime-only module comes
up under the mesh, serves its full tool surface with no valid credentials (the Servarr
lesson), stays up, binds its serve queues, and gets its scoped broker account.
SUITE_EXIT=0.
Claude-Session: https://claude.ai/code/session_01LrgweAeERJYBg88c5cKDzF
Prove gitlab — the exemplar tools-only, outbound-only external-SaaS
integration — installs: a scenario assigning gitlab to one node, and a test
asserting the mesh-runtime-gitlab container comes up and stays up, logs
[mesh-tools] serving 23 tool(s), binds its serve queues on the broker, and
gets its own scoped account — all with NO valid GitLab token, the case the
Servarr lesson is about.
No gitlab arm is needed in build-module-runtime.sh: gitlab speaks HTTP and
needs no extra CLI in the image.
Claude-Session: https://claude.ai/code/session_01LrgweAeERJYBg88c5cKDzF
A bed that assigns mosquitto and asserts the run-once step seeded dynsec
before the broker: the bootstrap ran to completion (not left running), the
seed is on disk owned by the broker's uid, the broker is up and stable
(it crash-loops against an unseeded store, so a stable broker is the proof),
and the node reached current. On top, the seeded admin authenticates over
MQTT and the provisioner grants a scoped client a consumer connects with.
build-module-runtime.sh gains a mosquitto arm (install mosquitto_ctrl from
the mosquitto package — it is not in mosquitto-clients on bookworm, and a
musl binary from eclipse-mosquitto would not load) and compiles the module's
bootstrap/index.ts entrypoint.
Claude-Session: https://claude.ai/code/session_01LrgweAeERJYBg88c5cKDzF
sonarr and radarr both access /services/media/downloads — the exact duplicate path
the resolver refused before novox/hq ADR 0051 (04-ISSUES/036, 012). Each now declares
it as an `access`, not a `directory` resource, so the pair co-resolves and one push
configures both. The operator provides the shared media dirs before apply (the host
refuses an absent access); the bed creates them after enrol and before the push.
Proves: the push is not refused, the node converges once, both modules' server and
runtime containers are up, and both server containers mount the same operator-owned
spool.
Claude-Session: https://claude.ai/code/session_01LrgweAeERJYBg88c5cKDzF
Adds the mongodb runtime CLI (mongosh) to build-module-runtime.sh, a
catalogue-apps scenario, and its install test. Proven so far: the ADR-0054 slug
applies and the mesh accepts the push (mongodb consumer identity mesh_anchor_mongo
fits). NOT green: the node applies but never reaches 'current' within 1200s — a
persistent reconcile divergence (applied-but-never-current, no crash), likely a
module declaring a resource its container mutates (issue-011 class). Needs live
VM inspection to name the module. Not merged.
Claude-Session: https://claude.ai/code/session_01LrgweAeERJYBg88c5cKDzF
Raises the first-node substrate and assigns postgres, minio, redis and
plex to one anchor in a single push, proving they resolve and come up
together on one node. postgres and minio each get a consumer that
connects with a real granted credential.
redis follows the corrected provider contract (ADR 0048, issue 032): its
runtime reconciles the contributions the mesh delivers at MESH_RECEIVES
and creates each consumer's ACL user with the mesh-minted password,
sealing nothing — no MESH_SEAL_KEY, no *.grant.json/*.credential path.
The provisioning proof authenticates as the consumer with the mesh's
password (PONG), matching the green provider-uses-mesh-credential bed.
Claude-Session: https://claude.ai/code/session_01LrgweAeERJYBg88c5cKDzF
Mirrors the postgres bed for minio: a provider (runtime carries mc) + a consumer
requiring s3-bucket, proving the consumer reaches its bucket with the access key and
secret the mesh delivered. It surfaced a real limit: the mesh derives `as` =
mesh_<node>_<module> (22 chars), and an S3 access key is capped at 20, so minio refuses
the service account. The test is correct and skipped pending 04-ISSUES/010, not worked
around.
Claude-Session: https://claude.ai/code/session_01LrgweAeERJYBg88c5cKDzF
Assigns a postgres provider (its runtime carries psql) and a module that requires
postgres-database; the mesh mints one password, postgres's provisioner creates a role
and database under the mesh's login with it, and the consumer connects to its database
with the delivered credential (a password-checked connection) — select 1. Nothing placed
by the test. The postgres half of the per-backend provider proof (ADR 0052/0053).
Claude-Session: https://claude.ai/code/session_01LrgweAeERJYBg88c5cKDzF
assigned-sonarr proves the Servarr detection path: the runtime discovers its API
key from the app's config.xml and serves its tools. assigned-grafana proves the
settings path: the operator states URL and token as settings, the mesh merges them
into the module's config file, and the runtime serves from that with nothing in the
manifest. Both green.
Claude-Session: https://claude.ai/code/session_01LrgweAeERJYBg88c5cKDzF
build-module-runtime.sh generalises the audit-logger runtime image to any module
(mesh-tools + sdk + the module's dist, entrypoints for tools/events/provisioner).
Two scenarios and two tests: assigned-plex proves a tools+events module serves its
tools over a mesh-issued scoped account; assigned-redis proves a provider's runtime
serves tools AND runs its provisioner in the same broker-bound process, provisioning
a grant and emitting its lifecycle event. Both green.
Claude-Session: https://claude.ai/code/session_01LrgweAeERJYBg88c5cKDzF
test/integration/assigned-audit.test.ts raises a node into a mesh, assigns it
the audit-logger through the control plane, and asserts the mesh delivered a
scoped amqps account (not the broker's own), the host ran the container, and an
emitted event reached the trail — the delivered credential authenticating is
the proof. scenarios/audit-node.yml is the lean single-node bed that stocks the
runtime image. Passes 1/1 against the real lab.
Claude-Session: https://claude.ai/code/session_01LrgweAeERJYBg88c5cKDzF
A consumer contributes a key prefix and gets an ACL user; the test is
that the grant means exactly what the manifest said, in both
directions: its own keys usable, anyone else's refused by the store
itself, and the flush a tenant must never have refused with them.
Waited for through the store rather than through logs: the user list,
asked with the password the host wrote into the server's own conf file
on the machine — nothing invented, both ends reading what the mesh
delivered.
And the forge is asked on the port the mesh assigned, not the one the
module declared. The old curl aimed at 3000, which was right until
ADR 0038 moved the machine side — a latent break that would have fired
on the first run to get past the settling that used to fail first.
The scenario stocks redis and its provisioner, and the rebuild builds
the provisioner image with the others.
Three changes, found by one failing test.
The forge failed three runs in a row as "status hangs", and it was
diagnosed twice as contention — real defects, fixed, and not the cause.
The heartbeats told the truth in the end: every exec on anchor crawled
from 15s to 105s, because eleven containers plus a database pull were
running in a 1GiB machine. Starvation presents as whatever you were
doing when the page-outs start, which is why it wore two other bugs'
clothes first.
So machine size is now the scenario's to declare — memory and cpus per
machine, default unchanged. The anchor that carries the whole substrate
is bigger than the laptop that joins it, and the comment on the
scenario says why in terms of what lands there.
`egress: true` gives a machine one extra interface on a lab-supplied
NAT network, addressed by DHCP because the one address a scenario has
no business choosing is on the host's side of the fence. Declared
per machine and off by default: a closed scenario stays the rule
(novox/hq ADR 0016), and the exception exists because a first node
fetches its images before any mesh can serve them — which is now the
tested path (04-ISSUES/029), and a lab that can never reach upstream
cannot prove the bootstrap it exists to prove. The uplink route is
metric-4096, so it never shadows a route the scenario declared. A
detached machine declaring egress is refused, not ignored.
And settled() treats a poll that threw as a poll that missed. An exec
timeout at minute four of a wait is "could not ask", not a verdict on
the machine.
**A password beginning with a dash broke the search for it.** The
credential test greps the machine's own files for the delivered
password; this run's password started `-S`, so grep read it as an option
and refused the whole invocation. The test compared the usage message
against "0" and reported the password as leaked. That is the worst way
for a search to fail — it says it found something. Fixed with `-e` and
`--`, which is what those exist for.
**The planning test could not redirect what the scenario does not
serve.** Rewriting an image reference only works for repositories the
scenario's registry actually holds, and the object store's provisioner
was not stocked — so that module kept its placeholder and the refusal
fired, correctly. It is stocked now, so all five are planned again. The
skip path stays for anything genuinely unserved, and says which module
and why: a planning test quietly covering four instead of five is the
false coverage this suite exists to prevent.
**The forge failed because of the one above it.** The planning test
threw before its cleanup could run, leaving a module assigned that
refused the next push, so no database container was ever created.
Yesterday's fix moved that cleanup where a failure cannot skip it — but
`after` still only unassigns what was assigned, so it now tracks what
actually got added rather than what was intended.
Everything up to now stopped at composing a declaration. That proves the
control plane and the host agree, and proves nothing about whether the
thing described works — which is how five modules sat pinned to images
that did not exist while parsing and resolving perfectly.
The forge is the right one to run first. It needs a database from
another module, a password it did not choose, and a connection string it
could not have written itself: the address and port come from what the
database serves, the user name from what the mesh decided both ends
would call it. If any of that is wrong it cannot start, and nothing else
in this suite would notice.
The test checks the chain in the order it has to happen — the login
exists, the database it owns exists, the forge answers, and its log does
not say authentication failed. That last one matters: a forge that
started and could not reach its database would still answer on its port.
Rewriting image references is now shared rather than copied from the
bundle, which had the same problem first. A digest is not knowable until
something is built, and when it is, it belongs to whichever registry
served it — so the text says which image and the scenario says which
copy. Matching is on the repository, with a test that a repository
ending in another one is not half-replaced.
Also makes the planning test put the machine back. Tests here share one
mesh, so the five modules it assigned were inherited by whatever ran
next; harmless while nothing pushed, and not harmless now.
**The adoption.** Software nobody here wrote, taking its credentials the
way such software does — from its environment — and needing two
containers that reach each other by name. The first module that could
not have been declared this morning: it needs the network shape and it
needs a sealed value to reach a container's environment.
Its password is accepted rather than generated, which is the whole shape
of an adoption: a service that already exists keeps the credential it
already has. Asserted properly — a wrong password is refused by the same
database, so the passing case means something.
**The warm scenario.** A mesh kept between runs and returned to, which
turned twelve minutes of bootstrap into thirty seconds of restore. Off
unless asked for: a run that is meant to mean something raises from
nothing.
Its guard fired for real during this work, unprompted — a mesh-host
commit landed and it refused the stale base, naming both commits, rather
than passing tests against yesterday's binary. That is 04-ISSUES/005's
rule one level down.
Three things the guard learned the hard way and now handles: a snapshot
captures disk and not memory, so the host is restarted after a restore
and asserted to have come back; the stocked image digests are worked out
while raising and a restored instance never raises, so they are kept;
and comparing only the repositories this run can see clears the ones it
cannot, so both directions are compared.
The one real bug behind five failed attempts was in mesh-host and it
reported itself precisely: a network shape the language had and no host
implemented. Everything else was scaffolding of mine.
The other half of the certificate split: the mesh's own authority
certifies internal names, and a name reachable from outside needs one
the world already trusts. Against a real server rather than a stub,
because what is under test is whether an order, a challenge and a
handshake agree, and a stub would be told to agree.
One assertion passes and one fails, and the failure is filed as
novox/hq 04-ISSUES/020: the authority issues a certificate and the
client never collects it. Kept as a failing test rather than deleted or
skipped — it is the reproduction, and it proves everything up to the
last hop.
The passing one is the guard that matters day to day: no certificate is
ordered for a name the mesh does not route, so a scan cannot spend an
account's rate limit.
The failure output gathers both sides before asserting. The first
version reported only what the proxy said, which made a server-side
question unanswerable — "the client never spoke to it" and "it refused
what the client said" are different faults with nothing in common.
Seven assertions against a real store, the important one being that a
consumer cannot reach another consumer's bucket — isolation here is a
policy somebody wrote rather than a boundary the product has.
The revocation test stages its own precondition. The first version
asserted a key left by an earlier test, and the rotation test had
already revoked it two tests early: the behaviour was correct and the
test was measuring residue. Its precondition assertion is what caught
that, rather than it passing green having verified nothing.
The request goes to the name, across the private network, and returns the
workload's own answer. Then the module is unassigned and the same request must
stop working — a stale public name pointing at nothing fails more visibly than
a stale grant.
The workload declares its port as well as its route, because they are different
questions and the earlier test leaves this machine filtering: a module that
asked for a route and not for the port would be unreachable by the proxy it
just asked for.
Two ends holding a matching string proves they agree, not that either is right.
So the check is three logins over the private network from the consumer's own
machine: the delivered credential works, the rotated one works, and the one
that was rotated away does not. Without the last, the test passes against a
provider that added a password without replacing one.
Not over loopback: pg_hba trusts anything there, and a deliberately wrong
password returned a row for a whole afternoon once.
Two setup faults, each of which looked like the thing being tested failing.
The builder module was assigned without its artifact ever being built, so
nothing could start — and the build has to happen while the hand-started
builder is still alive. Same chicken-and-egg as the registry, resolved the same
way: the builder that exists builds the one that replaces it.
The firewall test's listeners were squeezed through three levels of shell
quoting and never started, so the test failed on its own setup — which reads
exactly like the firewall working.
Written and loaded are different things, and loaded and enforcing are different
again. The test opens two ports on a machine, declares one of them, and checks
from the other machine that the declared one answers and the undeclared one
does not — then removes the module and checks the port closes with nobody
editing a rule.
The base image gains nftables, read back through `nft --version` like the other
three: a machine that cannot load a rule set applies the mesh's filtering,
reports success and filters nothing, which is the exact fault the derivation
exists to remove.
Two earlier tests were asking for things that are not there. The lab's registry
drops tags when it stocks, so `registry:2` is not served and the mirror test
failed with "not found" — it now uses the pinned digest, which is what a
declaration carries anyway.
The mesh generates a password, seals it to the machine that must accept
it, and discards the plaintext — so it cannot tell PostgreSQL to start
accepting it. Something on that machine reads what the host wrote and
makes it true. Everything up to that step is proven elsewhere; this is
where a password either becomes a login or does not.
A scenario with one machine and a database, and six assertions: the
password works, running again reaches the same state and says nothing,
rotation makes the new one work and the old one stop, a departed consumer
loses its login, a role nobody here made is left alone, and a manifest
naming a credential that was never written is refused rather than
creating a login with no password.
Each was confirmed to fail — and only it to fail — with the behaviour
removed from the provisioner: only-creates breaks rotation, no-revoke
breaks revocation, revoking everything breaks the bystander role, and
ignoring a missing credential breaks the refusal.
Two faults in the test itself, both worth recording:
- it checked logins from inside the database's own container over
127.0.0.1, which PostgreSQL's default pg_hba trusts. No password was
ever verified. Demonstrated directly: over loopback a deliberately
wrong password still returns a row. Only the rotation assertion
noticed, because it is the one that requires a password to STOP
working — which is an argument for writing that assertion every time.
- the fix then read .NetworkSettings.IPAddress, which docker 29 no
longer populates. It templates to empty, psql falls back to a unix
socket that is not there, and every login looks impossible rather than
misconfigured.
A sealed scenario cannot install wireguard-tools any more than it can install a
container runtime, so a lab without them cannot test connectivity at all --
which is most of what the mesh does between machines. Installed and not
started: what a node runs is the mesh's decision, and a lab that brought the
interface up itself would be testing its own setup.
growing-mesh exists to be grown. The point is not the third machine, it is that
adding one changes every other node's peer list -- so each has to be told
again, or the newcomer is on a network nobody else can see.
The first raises everything from the bundle its host carries and joins the mesh
it made; the second has a host and nothing else, and a person carries it a
token. This is the first scenario where the mesh is a mesh -- everything before
it proved a machine could talk to a control plane on its own loopback, which
proves less than it looks.
'incus list' failed because this shell had no permission to reach the daemon,
incusOk returned null, and the caller wrote ?? "[]". So 'mesh-lab list'
printed 'no scenario instances standing' -- confidently, about a question it
had never managed to ask.
The comment on incusOk warns about exactly this, in those words: absence and
success made indistinguishable. Three of its own callers then did it. Two
listings and the live diagram, which would have drawn an empty scenario rather
than fail -- a picture that is confidently wrong, which is worse than none.
Anything enumerating what exists now goes through enumerate() and throws.
incusOk stays right where failure genuinely means no, like instanceExists,
and there is a test holding that line so this does not get over-corrected
until nothing can be asked at all.
Worth noting 'mesh-lab check' already diagnoses this precise cause, down to
'a session that predates it cannot see it'. The diagnosis existed; the
listing just never asked for it.
One machine with PostgreSQL in its registry, for developing the bootstrap.
Used to verify that a sealed machine can raise a store and a database from the
bundle its host carries.
Closes 04-ISSUES/009. A scenario declares `images:` by tag; the lab stocks a
registry on this workstation where there is a network, raises it inside the
scenario as scenery, and reports the references a declaration pins -- which are
the digests THIS registry assigned, and are not knowable until it is raised.
Verified in a sealed machine, confirmed by ping to have no route out: package,
service including boot state, a container pinned by digest, and an action
inside that container. Applied, idempotent on re-apply, and read back from the
machine rather than from the apply's own report. That is the first time the
container shape has worked in the lab at all, and it was the shape blocking the
substrate bootstrap.
Four faults found by running it, three of them mine and one worth keeping:
The read-back checked that the catalog endpoint answered, by looking for the
substring "repositories" -- which `{"repositories":[]}` also contains. So it
passed on a registry holding nothing, and the failure surfaced much later as a
container that could not be pulled. It now asks for each image's manifest BY
DIGEST, which is what a machine does.
A recursive push needs its destination to exist, or incus copies the source's
contents rather than the source. The data landed one directory too shallow and
the registry found nothing where it looks.
The registry writes its blobs as root through a bind mount, so the workstation
could not remove its own scratch directory afterwards. Whoever made the files
removes them -- the cleanup now runs in a container too. And a cleanup failure
no longer fails a raise that succeeded: the scenario is standing and usable,
and saying otherwise would be a false report.
The base image build did not verify that the runtime trusts the documentation
ranges as plain-HTTP registries. Writing the file is not the daemon honouring
it, and a base image that looks right fails much later, in a sealed scenario,
a long way from its cause. It is now read back from `docker info`.
The lab raised an underlay and put nothing on it: correct, and useless, because
the thing it exists to test did not exist. Tier 0 now does, so `place: [host]`
works and a raised scenario finally contains something.
The refusal narrows rather than disappearing. A scenario placing a host and a
substrate is told which half is missing, by name — not that `place:` is
unsupported when half of it now works.
Placement reads back rather than assuming. A file arriving is not a host
working, so the binary is run before it is trusted to answer questions, and what
it reports is read from the machine (ADR 0035). The binary comes from an
explicit path, because the declaration design leaves where artifacts come from
open and a search would harden into the answer by accident.
The integration test that matters is the one asserting the host reports the
MACHINE and not the workstation that placed it. A raised VM and this workstation
differ in every capability — root versus uid 1000, a clean init versus a
degraded one, no docker versus docker, no wireguard versus wg0 — so a host
reporting the wrong machine is obvious here and invisible anywhere else.
And the placed host independently confirms ADR 0031: overlay absent on a freshly
raised machine. The underlay suite already asserted that by looking for
wireguard interfaces; this is a second witness rather than the same check twice.
Two tests failed the moment placement worked, which is what they were for. They
defended "there is nothing to place yet" while that was true; the decision
changed, so they change with it rather than being deleted.
Gate: 75 unit, 20 integration.
The full topology now raises: four machines, three routers, a transit
router, six segments, in 35 seconds. Everything the declaration model can
express except `place`, which is refused because the node host it would
place does not exist yet.
Transit was a real gap, not a bug. The design says public networks are
unrelated and routed to each other, never bridged — and I built the
segments and never built the thing that routes between them, so three
public networks were islands and nothing crossed. A transit router now
holds an interface on every public segment, forwarding and no translation:
the closest thing the lab has to the internet, deliberately dumb.
Proven rather than asserted, by ping TTL across the raised topology:
within one segment ttl=64 no hops
across two unrelated public networks ttl=62 gateway + transit
multicast between public networks 0 replies
A flat internet would have shown ttl=64 and answered multicast — which
would let a node discover a peer it could never reach in production, and
report success. That is the fault the as-is layer records the mesh already
hitting with multicast name resolution.
inbound: deny is implemented as a host firewall on the machine, read back
after applying. A declared refusal that silently did not load leaves the
machine wide open, which looks exactly like a machine that is working.
Established and related traffic is accepted, so a defended machine can
still dial out rather than being a disconnected one.
Verified by running, all of it:
home -> devices (policy allow) reachable
devices -> home (policy deny) blocked
behind unforwardable NAT -> out reachable
in -> behind unforwardable NAT unreachable
inbound: deny, dialling out reachable
reaching a machine that denies inbound refused
The two routers differ exactly as declared: the forwardable one carries the
policy rule and no inbound drop, the unforwardable one carries `ct state
new drop` and no DNAT.
A gateway is the one implicit machine in a declaration — a scenario says a
segment sits behind one and never names the thing that serves it. This
materialises it.
A router is a container, not a virtual machine, because it is scenery
rather than something under test (hq ADR 0033). Verified before building
that a plain unprivileged container can do all of it: ip_forward and ipv6
forwarding settable, nftables masquerade accepted, and the conntrack
timeouts mapping_ttl depends on both writable. No privileged mode.
Verified by running, on a machine behind a household gateway reached from
one on a routable address:
home-server -> anchor 0% loss, through masquerade
anchor -> 192.168.1.135 (private, direct) unreachable
anchor -> 192.0.2.50:8080 (the GATEWAY) HTTP 200
The last line is the published-but-behind-NAT case research 004 says only
exists in production. It is now a 32-second scenario on a workstation.
Segments sharing a gateway declaration share ONE router — that is what a
VLAN-capable router is, and two routers sharing an external address would
not work anyway.
mapping_ttl is read back after setting rather than assumed. Those sysctls
are not on every kernel, and a scenario that declared an expiring mapping
and silently got a permanent one would be exactly the fault being built
against.
Four bugs found by running it, three of them the same fault — a failure
made invisible.
The router had no route to a package repository, by design, so installing
nftables at raise time could not work. The image is now built once with
temporary connectivity and cached; every scenario after that needs no
network. That failure was hidden behind `|| true`, which is why it took a
raise to find.
The builder then failed on DNS: exec works before a container has an
address, and I had treated usable as ready. It now waits for the thing
actually needed.
The stock Alpine image ships `auto eth0 / inet dhcp` and its boot-time
networking service flushed the static address the scenario set — on eth0
only, so the outside interface came up bare while inside ones were fine.
The image build now neutralises it: a router reconfiguring itself from an
image default is the lab overriding the declaration. `ip addr add … || true`
had hidden this too, and is now `ip addr replace` with no swallow.
And routers were orphaned by destroy, holding their networks open so
destroy reported removing zero segments. They now carry the same machine
tag as everything else, so one query finds an instance's resources.
A declaration goes in and a disposable mesh comes out. Verified on a
workstation, not asserted: two machines raised and addressed in 14.6s,
snapshot 0.28s, restore-to-usable 11.6s, both families pinging with no
loss, and the workstation with no route into any of it.
The declaration layer implements the model in full — three positions a
machine can be in, keyed on forwardability; gateways carrying the address
the world sees them as; both address families; multi-homing; MTU;
inter-segment policy. It is validated hard because the failures it prevents
are silent: a private range on a public segment produces no error, the mesh
simply never forms. Public segments are refused unless they use RFC 5737 or
RFC 3849 space, and a range wider than the reserved block is refused too.
33 tests, all offline.
The runtime implements less than the model, and refuses the difference.
A scenario declaring gateways, published ports, policy, inbound deny or
place is rejected at raise with every gap named. Raising it would produce a
mesh that silently lacks what it declared, which is the fault this lab
exists to catch — 04-ISSUES/003, where a firewall key is declared in five
manifests and read by no code.
Three bugs found by review and by running it, all of one family:
The readiness check truthiness-tested incusOk's return. `exec … true`
succeeds with EMPTY output, so every machine reported unreachable while
incus exec on it worked perfectly. succeeds() now exists so the mistake is
not available, and network delete had the same bug — it counted zero
segments removed while removing them.
list() split instance from machine on the last dash, so a machine called
home-server absorbed half the instance id and destroy found nothing.
Resources are now found by the metadata they carry, never by name.
restore reported success in 0.79s while the machine's agent was still
starting, so the next command failed. Both raise and restore now wait for
usable and say how long that took — reporting the earlier number is
transport reported as effect, which is the fault the lab is being built to
find.
Two incus behaviours worth recording. Its CLI reads a YAML definition from
stdin when stdin is not a terminal, so a spawned command hangs until the
timeout kills it and arrives with empty stderr — a failure with no
explanation, on a command that works when typed. And it assigns a MAC at
runtime without recording it in device config, so MACs are derived and set
explicitly, which the guest needs anyway: it names interfaces by bus
position, and matching by name configures the wrong one on a multi-homed
machine.
No build step; Node strips the types. The lifecycle has no unit tests
because a fake hypervisor would assert that the fake behaves as expected,
which is the shape of test this project exists to stop shipping.
jschoubben