Context Agent 2.10.0 (bumped in #1219) moved to a multi-stage image whose
runtime stage ships no curl ("no vim, no git, no curl, no pipx/poetry",
nextcloud/context_agent Dockerfile at v2.10.0). The curl healthcheck then
fails with `exec: "curl": executable file not found in $PATH`, the task
never leaves `starting`, and the swarm converge wait gives up on
nextcloud_context_agent (main run 40401200083, Nextcloud#2).
Switch to the http_status flavor, which speaks HTTP over bash's
/dev/tcp and still requires a 2xx/3xx answer from /heartbeat.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Proxy:
- server_tokens off, X-Powered-By stripped from upstream responses,
HSTS max-age raised to one year.
- No global X-Content-Type-Options: under nosniff, ORB (step 3.4)
blocks no-cors cross-origin responses typed text/html, e.g. tracking
beacons answered with an empty 204.
PHP (CVE-2026-109618, CVE-2026-109619):
- New role sys-svc-compose-php installs a shared ini (expose_php Off,
session.upload_progress.enabled 0, fatal_error_backtraces 0 for
PHP 8.5, interim memory_limit -1) and a PHP-FPM [www] drop-in
(pm.max_requests = PHP_FPM_MAX_REQUESTS) once per stack host,
before 04_files renders any compose file.
- The host file names carry a sha1 prefix of the controller-side
source, so a content change alters every PHP app's compose.yml and
recreates it, also on remote deploys where compose_volumes cannot
read the stack host file and hashes the path instead.
- 20 PHP roles mount the ini as a compose/swarm config into their
image's scan dir; moodle, nextcloud and shopware also mount the FPM
drop-in. Magento keeps its own pool: its php-fpm.conf has no
include= and is rendered without notify.
- Joomla's php-upload.ini now loads as zzz-joomla-uploads.ini so its
own memory_limit wins: the media API compares Content-Length with
toBytes(memory_limit) and rejects every upload at -1.
TODOs record the memory_limit revert condition and accepted
trade-offs, Akaunting on PHP 8.1, FusionDirectory on PHP 7.4 and the
public Keycloak admin console.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Nextcloud#3 of this PR (run 38899717331, job 116977335535) failed
`tests/addons/integration_openproject.spec.js:30` in both retries: the
OpenProject admin section never rendered. The deploy had already logged
why: `occ app:install integration_openproject` answered `App "OpenProject
Integration" cannot be installed because it is not compatible with this
version of the server`, and the non-mandatory plugin was skipped while
the addon's `required: true` still sent the spec in.
Upstream, the latest stable release v3.2.0 declares
`<nextcloud min-version="33" max-version="34" />`; only the v3.3.0
nightlies (latest 2026-10-15) declare `max-version="35"`.
Keep `34-fpm-alpine` and mark the line `# nocheck: docker-version`, which
`utils/update/docker.py:suppressed_services` reads, so the updater stops
reopening the 35 bump. The nginx and context_agent bumps of this PR stay.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The role writes onlyoffice's settings with one occ call per key,
DocumentServerUrl first and jwt_secret fourth. Nextcloud loads app
config once per process, so a cron.php that starts inside that batch
runs onlyoffice's EditorsCheck job, registered when the app was
enabled, with the document server URL but without the secret. The
document server answers "Invalid token", EditorsCheck stores it in
settings_error, and onlyoffice stops loading its Files integration;
EditorsCheck itself skips every later run while that error is set.
Seen in run 38455782184 (Nextcloud#4, swarm): cron.php started at
06:50:00 inside the batch that ran from 06:49:59 to 06:50:04, logged
"Invalid token" at 06:50:22, and both Playwright passes found .docx
files offered for download only.
DocumentServerUrl is now written last, so a process that cached the
config before the batch ended sees no URL and EditorsCheck returns
early. After the batch, `occ onlyoffice:documentserver --check` sets
settings_error from a fresh check: a redeploy clears an error latched
earlier, and a document server that stays unreachable fails the
deploy. A unit test keeps the URL last.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Talk 25, the release for Nextcloud 35, reads stun_servers and
turn_servers through the typed app config and declares both as arrays.
Talk 24 read them as JSON strings. The role wrote them without a type,
so on Nextcloud 35 the Talk settings showed no STUN and no TURN server
although the deploy had set both.
A plugin configuration entry may now carry a type, which is passed to
occ config:app:set. The three Talk server lists use it. Entries without
a type are written as before.
Not yet run against a live Nextcloud 35.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Nextcloud 35 has no app store release of BPM Files and News yet, and the
role skips an addon it cannot install. Their tests ran anyway and failed
on the missing app. Recognize installs on 35, but its test trips over a
second element with the id it looks for.
An addon test runs only for an addon that is enabled and required. With
required set to false the three addons are still installed whenever a
compatible release exists, and their tests no longer run. metadata is
declared the same way already.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
A variant that pins `services.tor.enabled: true` while enabling a role whose
own `meta/services.yml` pins `tor.enabled: false` declares a combination the
deployment cannot hold. The role rides no node onion, so on that round its
surface lives at a clearnet host no onion browser reaches, and the dashboard's
`frame-src 'self' http://*.<onion>` forbids embedding it outright. Run
36651124458 job 109692287333 shows the result: five specs fail at once,
dashboard tile, matomo tracker, css injection, logout and the coturn metric
label, all on hosts that are the only non-onion ones in the whole job.
web-app-nextcloud variant 11 was such a round: bigbluebutton on, tor on, while
bigbluebutton declares no onion. Variant 8 already answers the identical
question for PeerTube (65538c40da, "clearnet, because the OIDC plugin rejects a
plaintext onion issuer"), so variant 11 now does the same. With that,
bigbluebutton never meets a tor node and needs no consumer gates of its own.
Two roles keep a gate, for reasons that differ from each other:
web-app-seaweedfs is not clearnet-only. It deploys on tor as the object store
other roles reach over the internal network, and only its browser frontend is
off there, which `frontend.enabled` on main already says. Its logout, dashboard
and prometheus declarations claimed a surface that round does not deploy, so
they take the same gate the file's own frontend and sso blocks use.
web-svc-coturn belongs on a tor node and cannot serve one: STUN/TURN runs on
UDP plus a 20000-port relay range that no onion descriptor can publish, and its
only HTTP is a gateway answering a plaintext OK. Deploying it is right,
advertising it as a prometheus consumer is not, because that scrape goes
through the browser path. It is exempt from the rule itself, since it claims no
visitable surface, and its `dynamic-flag` markers now carry the reasons the
rule's own message asks for; one of the four was a duplicate of the line below
it and is gone.
The lint states the rule. It found these five pairings on its own, matching a
hand count made before it existed, and reports none after the variant fix.
`test_sso_variant_requires_tor` covers the complementary half, every variant of
a role that has an onion must pin it on, and exempts the clearnet-only role
itself; this rule covers the consumers that bind one. The suppression marker
belongs on the bound role when the reason is that role's own shape, because a
consumer-side marker would also silence every other clearnet role in the file.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
PHP_MEMORY_LIMIT and PHP_OPCACHE_MEMORY_CONSUMPTION were both the host's
RAM divided by 30, read from two keys in meta/services.yml: 533M on a
16 GB host, 2133M on a 64 GB one, beside a container limited to 3g either
way. They are now half and an eighth of that limit through
lookup('service_memory_mb'), which renders 1536M and 384M on every host.
The lookup sits in env.j2 and the two meta keys are gone. It reads the
applications payload, and meta/services.yml is part of that payload, so
a call from there would run back into the render that asked for it.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The version-source updater only read meta/services.yml. An addon pinned in
meta/addons/<id>.yml stayed where it was: the ONLYOFFICE connector for
Nextcloud sat on v10.1.2, which does not install on Nextcloud 35, until
someone bumped it by hand.
The updater now reads addon files too. An addon declares its source in the
`update:` block it already has, beside monitored, catalog and upstream_id.
A monitored addon of the github-releases catalog needs nothing more: its
upstream_id names the repository whose tags are its versions. The
update-version-sources job opens the PRs, one per role.
Every pinned addon is wired:
- nextcloud onlyoffice carries its own version pin now. Its archive URL
stays a literal that carries the pin, because
test_pinned_addon_archives_available reads it as written
- wordpress mcp-adapter, moodle auth_oidc and jellyfin sso-authentication
follow the tags of their repository, moodle within its v4.5 series
- the three xwiki extensions read their Maven metadata
- pretix-oidc reads the PyPI simple index. The JSON API answers 404 behind
the package cache the test stack routes pypi.org through
- joomla's plugin is built here and jellyfin's ldap-authentication is bound
to the md5 of its archive; both are opted out with that reason
A bump rewrites the pin and the config.archive line together, and an
archive that does not carry the pin fails test_version_keys, so the two
cannot drift apart again. test_version_pins_are_watched covers addon pins
as well: a new one without a source or a reasoned opt-out fails the lint.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Connector v10.1.2 declares Nextcloud 33 to 34 as its supported range, so
occ refuses to enable it on Nextcloud 35 and the ONLYOFFICE variant of the
Nextcloud 35 bump fails in the onlyoffice addon spec. v10.2.1 declares 33
to 35 and therefore serves the pinned Nextcloud 34 as well as the pending
bump.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
ONLYOFFICE listed onlyoffice 10.2.0 in the Nextcloud appstore on 2026-09-18 at
10:16:31Z and never published the matching GitHub artifact; the tag itself does
not exist. `occ app:install` takes the newest compatible release, so it dies on
a 404 that no retry can heal. CI proved that twice independently: upstream run
35329488812 and fork run 35330645276 both fail Nextcloud#4 at that task after
six attempts, on the same commit, for the same reason.
The fault had three layers, each visible only once the one above it was gone.
Install. meta/addons/onlyoffice.yml pins `config.archive` to v10.1.2, whose
artifact answers 200 and which the appstore also lists for this server major.
01_install.yml stages that archive on the node and unpacks it into custom_apps
through a stdin pipe: the webroot is a named volume, so the node can only reach
the app directory through the container, and piping avoids both `container cp`
and a cleanup step that would have to run as a user which does not own the file.
Enable and configure needed no change at all. With the release already on disk
`occ app:install` reports "already installed", which nextcloud_install_status
has always mapped to runnable, so 02_enable.yml and 03_configure.yml proceed as
they always did. That is why this commit adds no branch to the plugin pipeline.
Update. `occ app:update --all` then dragged the pinned app back to the broken
10.2.0 and killed the deploy. occ offers no exclusion flag, but it does offer
`--showonly` and single app ids, so a pinned app is simply not updated. The old
`--all` call stays byte-identical and now carries a `when`: with nothing pinned,
thirteen of the fourteen variants take exactly the path they took before. This
layer cannot appear in CI at all, because it needs a second deploy with the app
already installed, and the install always died first.
The classifier test is inverted rather than deleted:
test_an_advertised_release_whose_artifact_is_gone_stays_fatal pins the decision
that a dead artifact fails loudly instead of degrading to a silent skip.
Verified by a local compose deploy of variant 4: ok=2508 changed=149 failed=0
and `✓ 36 tests/addons/onlyoffice.spec.js:12:1`, which drives the whole
JWT-authenticated round trip to the document server over its onion address, not
merely an enabled plugin. In the live container beforehand: `occ app:enable`
returned "onlyoffice 10.1.2 enabled", `occ app:install` returned "already
installed", and `occ app:update onlyoffice` reproduced the 404 that the pin now
avoids.
The loop expression was wrong before this commit and is fixed here: Jinja reads
'\1' as an octal escape, so the backreference produced chr(1) instead of an app
id. Evaluated against ansible-core's own filters over four shapes, the tail is
now cut instead of captured, which removes the escaping hazard entirely.
occ output is not localised: with default_language=de and a de_DE.UTF-8 locale
the strings are unchanged, so parsing them is as safe as the five markers the
role already relied on.
Residual: the per-app update branch has never run with a real item inside a
deploy. Its chain and `occ app:update <id>` are each verified separately.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The ranking the last commits built toward is now enforced: a semver tag,
else a digest, and nothing else without a written reason. Two states fail.
A pin that is neither names something the registry can move underneath us,
and nothing here can tell whether it still points where it did when it was
written. A digest on an upstream that does publish semver takes the
fallback for no gain and hides the entry from the update job.
Thirteen entries stood in the way when the check was first run. Four were
resolved by the channel grammar, which made the Jitsi components orderable
without touching their pins. Three needed nothing once a moving tag could
declare itself with 'docker-version'. Two are fixed here and four are
declared.
LiteLLM leaves 'main-v1.77.3.dynamic_rates'. That tag exists exactly once
among 2609, so it is a dead end no updater can follow, and it misreports
itself: the image inside answers 1.77.2. The pin moves to
'main-v1.77.7-stable', five patches along the same minor and onto a channel
with 134 builds. Started against this repository's config shape, 1.77.7
registers both model entry forms and accepts drop_params and the
os.environ master key.
Pixelfed leaves 'latest' for '20260917'. The upstream stamps its dates as
one eight-digit number, which is depth 1 and orderable with no tooling
change at all, and the two tags resolve to the same digest, so nothing
about the deployed image moves.
The four declarations carry the count that justifies them rather than a
claim: LM Studio publishes two tags and neither is a version, the MIG
registry carries one, and the two Nextcloud Talk images publish 174 and 104
AIO build stamps between them.
Both branches of the gate were negative-controlled. The digest branch
skips in this repository because no digest pin exists at all, so it was
exercised by introducing one and watching it fail.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
API is a plain mapping, so an inventory that configures three providers replaces the whole declaration from group_vars/all/18_api.yml instead of merging into it. inventories/development/default.yml declares openai, anthropic and openrouter, so github, google and jira did not exist in CI at all and lookup('api', 'github.client_id') raised.
A raising lookup inside a value the applications renderer is templating does not fail the render: _templar_render_best_effort returns the string untouched and, for a value carrying a non-env lookup(, refuses the fallback as well. The raw template text reached plugin_value.enabled | bool, which coerced it to False and emitted the deprecation warning that fails the deploy gate. Run 34893363518, job 104153194109 lost web-app-nextcloud to exactly that, for three integrations at once. The configvalue entries would have stored template text as the client id.
utils/api.py merges the scope's override over the declaration, so an override narrows values and never the provider set, and holds the enable rule: every credential a provider declares must carry a value. The api lookup reads through it; the new api_enabled lookup gates on it and returns a real boolean instead of a string a filter has to coerce.
The same class sat in group_vars/all/16_ai.yml and roles/svc-ai-litellm/meta/secrets.yml as bare API.<provider>.<key> dereferences; both now go through the lookups.
Two lints keep it shut. test_addon_enabled_renders_to_bool renders the whole applications map against a scope with API blanked, the most hostile override an inventory may legally write, and requires every addon gate to come out a boolean. test_api_accessed_via_lookup forbids a bare API dereference in Jinja under roles/, group_vars/ and inventories/.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Fourteen roles repeated the same two lines to say they use the gateway: a
`shared` flag gated on group_names, and an `enabled` flag conjoining that gate
with LITELLM_CHAT_MODEL_SERVED. Both are now single-sourced.
`shared` becomes literal true. The flag means the provider is a shared instance
rather than one deployed per consumer, and a central gateway is always that;
the dynamic form only said "shared if deployed", which `enabled` already
decides. The 50 shared pins under `litellm:` leave meta/variants.yml with it.
The tree had already noticed: 20 of those 50 lines carried
`# nocheck: variants-static-override`, two of them with copy-pasted text about
container_backup. This narrows e5de4e085a for one service key rather than
reversing it; every other dynamic flag keeps the group_names form.
`enabled` becomes `{{ LITELLM_USABLE }}`, defined in group_vars/all/16_ai.yml.
That shorthand is only possible because is_explicit_truth learns about it.
The resolver reads a dependency out of the flag TEXT — "in group_names" in
value — so a constant hides the role id and the provider silently leaves the
static closure of every role referencing it. Measured: the old expression
resolves True, `{{ LITELLM_USABLE }}` resolves False. GROUP_GATED_CONSTANTS
names the constants whose own definition carries the gate, and the docstring
states the obligation: a SPOT of this kind that is not listed there drops its
provider without a word.
The wider fix, accepting any Jinja as dynamic, was measured and rejected: 35
other flags carry a Jinja value without the marker (MCP sidecars, javascript
and SSO gates, recaptcha), and they would all enter closures they are not in
today.
A literal shared flag is no longer dynamic, so test_dynamic_flags wants the
`# nocheck: dynamic-flag` its own message documents for legitimate exceptions;
where a marker already existed the rule list is extended rather than a second
comment added.
Behaviour is unchanged where it counts. The closure keeps litellm with the same
app counts (wordpress 26, openwebui 23, hermes 23, mediawiki 22, flowise 23),
unrelated Jinja still resolves False, and the truth table of the new enabled
form matches the old one in all four served/deployed combinations. Three roles
resolve without litellm, and `git show HEAD:` proves that predates this change:
zammad pins enabled false in variant 0 at line 39, nextcloud at line 97.
Six unit tests cover the resolver, including one asserting that every
registered constant is actually defined in group_vars, so an entry cannot
become a dead promise.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Ansible renders the values of a structure loaded through include_vars and leaves its keys as written, so the mapping key a564385068 wrote as a Jinja expression reached Nextcloud as its own source text. occ config:app:get sociallogin custom_providers on a live variant-1 instance returns "groupMapping": {"/{{ lookup('rbac_group_path', application_id=application_id, role='administrator') }}": "admin"} while every neighbouring value of the same provider is rendered, scope among them.
Keycloak emits /roles/web-app-nextcloud/administrator, so the mapping matched nothing and the account sociallogin provisions on first login stayed out of admin. occ group:list on that instance shows admin holding only the local administrator while keycloak-ac165525-e9e8-4230-9381-13b1acd83fc0 exists and belongs to no group, and every Playwright spec that needs an admin settings section answered 'Access forbidden'.
An independent reproduction outside this repo showed the same split on ansible-core 2.21.4 and 2.20.6: in a payload of the same shape the sibling value rendered and the key did not, so it is not a version artifact. Nextcloud 34.0.3 settles the other half: Manager.php hands an admin the unfiltered settings of a section and ai is a core section of the always-enabled settings app, so an admin cannot receive 403 there whatever is installed.
Writing the whole mapping as one expression makes jinja2_native, which this deployment has on, return a real mapping with the key computed. The lint takes the class: no meta payload may carry Jinja in a mapping key. It found exactly this one case in the whole meta tree, and it ignores a line that itself opens with the expression so the corrected form does not trip it.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Sociallogin creates a Nextcloud user as "<provider name>-<keycloak sub>",
and the provider name was the Keycloak domain. On an onion variant that
domain alone is 67 chars, so every OIDC callback died with
InvalidArgumentException "Username is too long" and HTTP 500 at
/apps/sociallogin/custom_oidc/auth.<onion> (run 34615516009,
Nextcloud#1, 204 occurrences). The whiteboard and admin OIDC specs failed
on the same login.
This is the second derived identifier an onion domain overflows; the
kernel hostname is the first, clamped to 63 bytes by container_hostname
since the pretix and bookwyrm fixes. The domain-or-entity rule now lives in
one lookup, bounded_name(application_id, max_length): the primary domain
when it fits, otherwise the entity name. container_hostname delegates to it
at 63 and keeps its eight callers and its behaviour.
The provider name is bounded_name('web-app-keycloak', 64 - 1 - 36), the uid
limit minus the separator and the UUID sub. A short clearnet domain stays
the name, so existing users keep their uids; an onion deploy gets
"keycloak", which, unlike a hash of the domain, survives a rotated onion
address.
Tests cover the lookup's fit, boundary and fallback cases and its argument
validation, container_hostname's delegation at 63, and the provider name's
room for the sub.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
start_interval is the cadence while start_period runs, and only while it runs.
Once the window closes the probe falls back to interval, so a service that
becomes ready just after it pays a whole tick anyway and the cadence never
reached it. Six of the seven flavors were shaped that way, Curl worst: a thirty
second window against a one minute interval.
Measured across 151 converge waits in four job logs, 63 sit exactly on a tick
boundary, 36 in the thirty second step and 27 in the sixty second one.
Attributed by application, web-app-seaweedfs is the largest at 1501s over
sixteen waits with a median of 57.7s: it probes with curl and fell straight into
the gap. web-app-zammad probes with connect, whose window is fifteen minutes,
and its 154s median is real boot time with nothing to recover.
So the window moves to ten minutes and the steady state interval to two, and the
flavor overrides that sat below them go: Curl, Wget and Connect lose interval
1m, Http its twenty second window, Nc its ten. Connect keeps fifteen minutes.
Neither number costs anything while a container works. start_period delays
nothing, it only stops failures counting, and interval no longer touches
convergence at all since that hangs on the window alone; it is steady state
load, which drops from 120 probes an hour per service to 30.
Forty six role files follow. Twenty seven declarations go that repeat the new
default or undercut it, an undercut window being the defect itself. Eight
windows longer than ten minutes keep their value and now say why. The literal
blocks and the three Dockerfiles this repository builds spell every timing out,
because outside the lookup an omission takes docker's value and not the
platform's.
A fourth lint rule keeps the shape: start_period must outlast one interval,
checked against the effective values, so raising interval without the window
cannot quietly bring the cliff back.
Two faults in the lint surfaced while building it and are fixed here. It
compared timings as strings, so 600s and 10m read as different values, and it
matched the healthcheck key exactly, so a line carrying a trailing nocheck was
skipped whole, which is how the two swarm-init partials went unexamined.
One regression, bounded. A container that starts and never turns healthy is
neither rejected nor errored, so stack_ready.sh only calls it fatal after three
killed tasks, and each cycle now spans the longer window. That reaches twenty
minutes rather than six, capped there by swarm_stack_converge_retries at 600
with a two second delay, and it lands only on runs that were failing anyway.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
confluence 10.2.16 to 10.2.18, erpnext v16.34.1 to v16.34.2, espocrm 10.0.7 to 10.0.8, hermes v2026.7.20 to v2026.9.7, mattermost 11.10.1 to 11.11.0, nextcloud context_agent 2.7.0 to 2.8.0, seaweedfs 4.45 to 4.46, coturn 4.17.2 to 4.18.0. The automated updater opens one PR per role and had not reached these eight; playwright and elasticsearch were already current in the tree.
audit.md follows rather than leads: the MCP report carries a version column and test_mcp_audit_report compares the committed file against a fresh generation, so hermes and mattermost move there too. Regenerated with audit.gen.py as that test instructs. Note that make docs does not run this generator, so the report only tracks reality when somebody calls it by hand.
Verified by the docker-version check itself: eleven Outdated warnings before, zero after.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Nextcloud#2 of run 33923190220 probes the MCP surface three times and gets
"could not reach http://contextagentmcp:8080/mcp: <urlopen error [Errno 111]
Connection refused>" before failing on both workers. Connection refused, not
an unresolved name: the sidecar's address existed and nothing served it.
meta/services.yml derives both halves of that surface from the same switch,
context_agent and contextagentmcp each carrying
enabled: "{{ lookup('config', 'web-app-nextcloud', 'mcp.enabled') | bool }}".
mcp.enabled is itself derived, from peer presence rather than by hand:
utils/roles/applications/mcp.py renders "'peer' in group_names or ..." over
the consumers the role admits, and for nextcloud that is hermes, openclaw and
openwebui. Variant 2 is the openwebui variant, so the switch is true, the
sidecar comes up, and only the app behind it stayed down.
It stayed down because the variant pinned it off by hand. That pin is an
orphan. a2be4f081a introduced context_agent paired with an mcp pin in each
variant and kept the two in step; 06bc856c32 lifted mcp out of the services
map when it became a role topic; ffdef28da3 then dropped the hand-maintained
mcp values entirely, calling them "wrong in both directions", and replaced
them with the peer derivation. The mirrors it left behind have contradicted
that derivation ever since.
Removing them changes exactly one variant. Evaluated across all fourteen:
variants 0 and 1 keep their explicit enabled: true and are untouched, variant
2 gains the agent its sidecar needs, and variants 3 through 13 admit none of
the three peers, so the derivation yields false and they behave as before.
Eleven of the twelve removals were writing a value the derivation already
produced.
The budget objection does not hold either. collect_role_resources counts any
enabled that is not literally false, so a rendered template counts as on;
test_resource_budget therefore charges the agent's 512m to every variant that
no longer pins it off, and still passes.
merge/svc-net-tor still carries all fourteen declarations and will reintroduce
the orphans if it lands.
Nextcloud#1 of run 33923190220 fails two admin surfaces:
addons/whiteboard.spec.js:28 waiting for the settings/admin/whiteboard shell
and addons/integration_openai.spec.js:32 waiting for #openai_prefs on
settings/admin/ai. Both time out with "element(s) not found" after two
retries.
The cause is not the apps. In the sociallogin flavor the Keycloak login
provisions a second Nextcloud account whose uid is <provider>-<keycloak sub>,
while the role only ever puts the local administrator account into the admin
group, so the browser session that the specs drive is an ordinary user and
every /settings/admin/* route renders nothing. sso_oidc_plugin.py returns
"oidc_login" if ldap_enabled else "sociallogin", so this only bites where LDAP
is off: variants on oidc_login with oidc_login_proxy_ldap land on the
LDAP-backed account that 07_admin_identities.yml adds to admin instead.
The provider asked for scope "openid" with an empty groupsClaim, so no group
information reached the plugin at all and no mapping could have worked. It now
requests the RBAC scope, reads the claim, and maps the application's
administrator group onto Nextcloud's built-in admin group.
Every link is in place already. RBAC.GROUP.CLAIM is "groups", the Keycloak
client scope carries the same name and sits in defaultClientScopes, and its
mapper sets full.path true, so the claim carries /roles/<app>/<role> and the
leading slash in the mapping key matches. rbac_group_path accepts
'administrator' without a declaration because it is the implicit admin role,
and build_realm_rbac_groups adds that group to every RBAC application, so the
group the key names is really provisioned. Upstream reads both keys:
ProviderService maps groups_claim to groupsClaim and group_mapping to
groupMapping, and applies the mapping by group id.
Adminship is necessary for those two specs but may not be sufficient: both
assertions also claim their app is installed and enabled.
The gateway publishes only models it has a backend for and renders model_list: [] when neither ollama nor lmstudio is deployed and no OpenRouter key is set. LITELLM_CHAT_MODEL fell back to openrouter/auto without checking that key, so consumers were configured for a model the gateway does not carry and every prompt failed with a 400.
Observed in the Mattermost#1 CI job: litellm.log carries 12 occurrences of "Invalid model name passed in model=openrouter/auto" next to 400 Bad Request, and tests/test-litellm-agent.js failed on every attempt while the rest of the suite passed.
Make the fallback conditional on the key so the name is empty when nothing is served, promote LITELLM_CHAT_MODEL_SERVED from the role vars to group_vars where consumers can see it (it was the third copy of this logic and the only correct one), and switch the 14 consumers from "litellm is deployed" to "litellm serves a model". The Playwright service flag reads the same lookup, so the spec skips together with the Ansible gate. The bool filter carries weight: the folded scalar yields the string False, which Jinja treats as truthy.
A new unit test renders both expressions against Jinja and pins the invariant that a model name exists exactly when a model is served, including the case where ollama preloads embedding models only.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
plugins/lookup/tls.py:102 builds url.base as "{protocol}://{domain}/", pinned by test_tls.py:59. meta/addons/onlyoffice.yml appended another one, so DocumentServerInternalUrl was stored as https://host//.
The connector treats its two URLs differently: getDocumentServerUrl normalises with rtrim, getDocumentServerInternalUrl returns the stored string raw and concatenates the endpoint onto it. Every server-side call therefore went to //healthcheck and //converter. A swarm vhost forwards $request_uri verbatim, so the document server answered 404; a compose vhost uses proxy_pass with a URI part and nginx collapses the double slash. The 404 makes the connector cron write settings_error, and FilesListener then stops adding the onlyoffice script to the Files view, which is why iframe#onlyofficeFrame never appears and the addon spec times out. Observed in run 33828309745 job 100890535389: "GET //healthcheck HTTP/2.0" 404 in the openresty log, and the docx row offering Download as its default action.
The lint generalises it. This class has been fixed one site at a time at least five times before (efee415b21, 035c1efbf3, b0c7c9601e, a96bdbba38, 83636408c9), so the rule now holds repo-wide: a variable carrying an unstripped tls url.base may never be followed by a slash. Definitions that strip it again are excluded, which is what keeps KEYCLOAK_URL and OPENTALK_KEYCLOAK_BASE_URL out of the finding set; measured, no name is defined both ways.
Not fixed here: settings_error is sticky, so an instance already poisoned stays broken after a redeploy. The obvious remedy, gating the deploy on occ onlyoffice:documentserver --check, is a four-step round trip whose steps never succeeded in either artifact, so it needs its own design.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The hub serves 94 tools across 23 categories, write tools included, and enabling the ExApp switches every category on. That surface was published on the public vhost and every discovered client was handed an app password for it. svc-ai-mcp-adapter now sits in front in mcp_passthrough mode: clients present credentials.mcp_bearer to contextagentmcp and reach the nine read tools of calendar, contacts and files, which files/mcp/tools.json names and the sidecar enforces per call. Those three categories are the ones whose upstream is_available is unconditional, so the served set does not vary with which optional apps a deployment installs.
Enforcing it in the sidecar rather than in each client is what makes it hold: web-app-openwebui renders no per-tool filter at all, so a client-side include list would have bounded two consumers out of three.
06c13e7f3e withdrew the declared path from the edge for exactly this reason - "that surface is what the adapter exists to narrow". Nextcloud answers the same route under a second spelling, without the /index.php front controller, which try_files rewrites back onto the route after the edge has already declined to block it. The withdrawal becomes a regex covering both, and only where stripping the front controller changes the path, so the four adapter providers without one render unchanged.
The app password is minted after the stack renders, so it reaches the sidecar through upstream.env and rebuild.yml rather than the compose environment: compose up resolves env_file while creating the container, a restart does not. The sidecar is the last service in the template because a Jinja set has no block scope and its env_file would otherwise displace the one talk, whiteboard and talk_recording each carry.
Two suites cover what nothing measured: the vhost template rendered without a container, asserting both spellings are withdrawn, and every shipped tools.json loaded through the adapter own loader, so a bad contract fails in a second instead of as a probe timing out an hour into a deploy. The guest spec probed mcp.endpoint.path, which this change makes the proxy answer 404 for its own reasons; it now probes all three public paths and requires no MCP body.
Not yet deployed: the sidecar build, the upstream.env delivery and the hub answering exactly nine tools are unverified against a running stack.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
supported_version and minimum_version sat in 24 roles and were read by no template, no task and no lookup. Their only consumer was the audit report version column, which turned false the moment an automated image bump ran: qdrant reported a verified version nobody had verified, two releases long. Both leave MCP_KEYS, so a role that declares them again fails the schema lint instead of carrying a field with no effect.
Requirement 025 mandated minimum_version as a MUST and 035 carried it in an example; both are corrected here, otherwise the repository would contradict its own code.
The gate then surfaced five dead docs.mattermost.com links. Mattermost moved to /administration-guide/ and /deployment-guide/, and every replacement was probed for 200 rather than guessed. The MCP page is gone with no successor, so mattermost now points at mcpserver/README.md in the Agents repository, the source rather than a rendered mirror.
Both concerns share three files - mattermost's meta/mcp.yml, requirement 025, and the generated audit.md, which derives from a source of each - so splitting them would commit an artifact that does not match its sources.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
An MCP client spec signs in through the client's own OIDC button, which
redirects to Keycloak. Every provider put its MCP pairing in the lean variant,
and that variant pinned sso off - so the round that exists to prove the pairing
was the one round without an identity provider. Run 33574056480 shows what that
costs: openwebui's MCP specs reached _shared.js:131 and died on "openwebui OIDC
sign-in button must be visible on the auth page".
The rule now demands two things of the variant that carries the consumers: the
provider pins sso on, and each admitted consumer, in the variant that round
actually pulls in, does not pin it off. The second half matters because a round
deploys every dependency at its own index and falls back to variant 0 only when
the dependency has fewer, so the client shipping alongside is that one and not
the client's variant 0. A role that declares no sso, or pins it to a literal
false because upstream speaks no OIDC, is exempt from a flag it cannot have.
Twenty roles moved to satisfy it. Ten already had a second sso: false pin and
only needed the flip. Six could host the pairing in variant 0, which already
has sso on, and dropped back to two variants. Six kept a third lean variant to
carry the false polarity variants-coverage wants. moodle and mattermost took
swaps instead: moodle's variant 1 is the mandated LDAP-only branch, and
mattermost was already at the three-variant cap with sso on in its variant 2.
nextcloud also turns context_agent on in that round, the ExApp that serves the
endpoint - with it off the round proved nothing it was built to prove.
The rule moves to tests/lint/ansible/services/ because it reads meta YAML and
executes nothing, which is where integration.md says such a check belongs, and
its suppression key joins the catalog it was never listed in.
nextcloud_exapp_reregister becomes nextcloud_exapp_renew: mcp-secret-masking
matches `register:` unanchored, so the substring in the old name read as a task
that prints a secret. register.yml already masks the two tasks that carry one.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The mcp-consumer-variant guard demanded three things of a provider's MCP
round: every admitted consumer enabled, each of them pinned `services: null`
so it dragged no dependencies along, and no other service left enabled. Two
of those were the rule legislating a round's contents rather than checking
that the pairing gets proven at all.
Keep the first and drop the rest. What else the round carries is the role's
own call, so the tor/litellm exemption, the datastore exemption and the
`allowed` set they fed have no reader left and go with them.
Strip the 63 `services: null` pins the rule forced across 20 variants, all
written bare as `services:`. The topic-emptying capability behind them
stays: apply_topic still normalises None to {} for the inventory payload,
and the unit tests still cover it. Only the uses and the mandate are gone.
That leaves overridden_providers scanning a repository with nothing left to
find, so its three assertions passed over an empty dict and would have
stayed green however the function broke. Rebuild them on a synthetic role
tree, which also covers the case the repository scan never held: a variant
naming a plain flag is not an override.
Removing the pins changes what deploys. Measured across all 20 variants: 18
identical, prometheus#1 grows 23 to 31 roles, moodle#2 grows 31 to 32 and
crosses the memory budget until the following commit shrinks it back.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
A round deploys a provider once, so the role that pulls it in is the only place that can say how lean it has to be. A variant entry may now carry any application config topic under a service it depends on: a mapping deep-merges, so naming one key changes that key alone, and null empties the topic. Empty was the case that needed a marker of its own, because merging an empty mapping is a no-op and would leave exactly the services the round was trying to shed.
The topic vocabulary and the merge semantics live in utils.roles.applications.topics, which is also where resources.py now gets them rather than spelling the same null-or-merge out a second time. Collecting who demands what stays in the inventory tree, refuses two roles demanding different values for one provider in the same round, and skips a role's own entity entry so the existing services.<own>.users declarations are not read as overrides.
The CI set-cover learns the same fact. Its greedy cover retires a role's variant-0 row when an earlier row of another role embeds it, on the stated premise that embedding pulls the provider's variant 0. A variant that dictates that provider's config breaks the premise from the other side: whatever ran, it was not the provider as the provider declares it. Such a pair now covers nothing, so the retired row keeps its job, its artefact and its verdict.
Measured on the current tree: 20 override pairs, 41 covered rows before and after, so nothing is revoked today. The rule bites the moment an override lands in a row that covers.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The contract check ran as an Ansible task in the middle of every provider deploy, and the two client checks ran the same way. A test that is part of the deploy cannot be run without deploying, cannot be re-run on its own, and reports its verdict as a deploy failure. Twenty-seven task files lose 491 lines to this.
It is now what it always was. Each provider ships files/test/test.sh, templates/test.env.j2 and meta/tests.yml, and test-e2e-cli stages one shared harness beside it: run.sh drives contract.sh in a throwaway container on the provider's own network, then isolation.sh for adapter-fronted providers. The read probe each provider offers moved from the calling task into meta/mcp.yml next to the allowlist it must satisfy, and the schema now rejects a read probe naming a tool outside that allowlist.
The bearer the test presents is the one a client resolves through mcp.credential, via the new mcp_client_bearer lookup over resolve_credential, which moved to utils.roles.applications.mcp so it is the same function mcp_servers renders every client connection with. The probe presented whatever the calling role passed, which for the fifteen adapter-fronted providers is a different secret than any client holds.
Sidecar isolation is read from the engine rather than measured inside the container, so it no longer needs a delegation to the node holding the task: attachment is compared by network name, which drops the gwbridge special case, and a swarm worker says it cannot measure instead of skipping silently.
Retrying is bounded by wall clock against MCP_DEADLINE_SECONDS, which scales through the same lookup('timeout') as the harness budget it has to stay under. A flag that is neither true nor false aborts rather than reading as off.
mcp-provider-probe becomes mcp-provider-cli-test: an enabled provider must ship the test rather than include the task. mcp-probe-bearer becomes mcp-adapter-bearer, tying ADAPTER_BEARER to the declared credential key now that no probe include carries it. A new integration test renders the shared env template for every provider through Ansible's own templar, so a later change to it breaks visibly instead of poisoning twenty deploys quietly.
Hermes and OpenClaw keep their client-side proof as CLI tests of their own.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
6ac48b7b8 gave talk_recording an enabled, which is what makes
test_env_services_match look for a TALK_RECORDING_SERVICE_ENABLED line: an
entry without the key is invisible to that rule, one with it is not.
The container is reachable only from the Talk signaling backend and serves
no browser surface, so it takes the marker context_agent already carries
rather than a flag no spec could gate.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
A compose.yml.j2 that wraps a service key in {% if VAR | bool %} has to reach
that service's own meta/services.yml entry, or the template and the
declaration become two switches for one container: the resource model, the
README graph and the network layer read the entry while the container follows
something else.
The reference may sit anywhere in the chain. The resolver walks the role's
vars/ and defaults/ and follows config paths into the resolved application
config, because nextcloud reaches its Talk entry over
addons.spreed.enabled and a resolver that stopped at role variables called
that a bypass. Only guards indented by at most two spaces count: deeper ones
select content inside a service, not the service.
The rule found one real bypass. NEXTCLOUD_MCP_ENABLED read mcp.enabled while
switching the context_agent container, whose entry admitted two peers where
mcp.enabled admits three, so a round carrying only web-app-openwebui ran a
container its own entry declared off. mcp.endpoint.path routes through
app_api/proxy/context_agent/mcp/, so the container is the MCP surface and the
entry was the stale side. Both ends now meet at the entry.
talk_recording carried no enabled at all and fell back to
_looks_like_container, so the resource model billed it unconditionally. It
now reads services.talk.enabled, which is what its guard already computed.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
83c1c729de gave every provider a group_names-gated services entry per MCP
client, including the clients it refuses with mcp_consumer: false. The gate
is what makes the inventory resolver plan provider and client into one round,
so 19 of the repo's 82 co-deploy pairs started a client for a pairing the
provider had already declined. flowise is a live MCP client of qdrant,
homeassistant and prometheus; the other 17 providers were bringing up its
node stack for nothing.
The entries now read enabled: false, shared: false, mcp_consumer: false, and
the variants that pinned them to true are gone. Setting those variants to
false instead is not an option: test_variants_no_static_override forbids any
variant override of a literal base value in either direction.
bond leaves with them, and a new lint holds that line: collect_role_resources
skips a literally disabled entry before it reads any resource key, so a bond
there describes a cost nobody pays. Thirteen entries unrelated to MCP carried
one and are cleared too.
test_mcp_consumer_services_entry demanded the group_names shape for every
client entry, refused ones included, and now demands the opposite there.
dynamic-flag keeps its rule and takes its own documented per-key exemption.
make docs redraws twenty graphs: the edges go, the nodes stay as
'openwebui ❌' and 'flowise ❌', so the refusal is visible rather than read as
an integration.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
enabled and shared were hand-maintained in 24 roles and wrong in both
directions. Both now resolve from the mcp_consumer edges already declared
in meta/services.yml: a server takes the consumers it admits, a client
takes the servers that admit it, and the pair renders as "'peer' in
group_names or ...", or False when no peer is admitted at all. setdefault
keeps meta/mcp.yml and meta/variants.yml as the override channel, so
pinning either key still wins.
The resolved block replaces the mapping instead of being written into it.
_build_role_base_config parks the YAML cache's own object under mcp, so an
in-place setdefault injected the derived keys into every later raw read and
made test_mcp_schema pass alone but fail in a directory run.
The derivation keys on peer presence, never on a peer's own value: a
re-entrant read under _RENDER_GUARD returns the unrendered dict.
Four lints now ask the resolver instead of reading meta/mcp.yml, which they
could not do while the key was hand-written. That is what exposed the eleven
missing hostile-path specs the previous commit closed.
Carries one change that belongs to the WordPress probe but shares a file
with this one: the tool allowlist moves to the hyphenated names the adapter
actually serves.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
allowed_consumers was twenty hand-kept lists of which fourteen were identical, so every new client meant an edit in every provider. A client now declares services.<own-entity>.mcp_consumer: true once and a provider refuses one with mcp_consumer: false, which is also what tells a deliberate exclusion apart from a forgotten entry.
The group_names-gated consumer entries give the inventory resolver the co-deploy edge meta/mcp.yml never had: one of sixty-nine matrix rounds carried an MCP client before, twenty-three role/round pairs do now. bond 0.5 keeps the clients out of round 0 so the host memory budget still holds.
BREAKING CHANGE: meta/mcp.yml no longer accepts allowed_consumers; declare services.<entity>.mcp_consumer on the client instead.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
