The deploy decides what ADAPTER_PORT or OPENWEBUI_ADMIN_NAME is and
writes it into the environment. A default in the reader is a second place
that value lives, and it is the one nobody greps for. os.environ["X"]
fails loudly on a missing key, which is the behaviour a misconfiguration
deserves.
Empty values stay free, because they add no second source and only say
absent: "", None, False, 0, the empty containers, and the "[]"/"{}" a
json.loads reader hands an absent variable. A default that is itself an
environ.get is free too, since it reads the one source again under
another key; a literal ending such a chain is still reported. Detection
runs on the parsed tree, so a call split across lines reads the same.
Five sites were real duplication and lost their default: the adapter
port, openwebui's admin name (a role that runs as a bot user), odoo's
OIDC uid field, seaweedfs' public-primary flag, and the swarm hosts
entry IP. Nineteen carry a nocheck naming why the value is not the
deploy's: contributor tooling whose default is the only source, test
scaffolding that writes the file it names, and two readers of
CA_TRUST_CERT_EXTRA_HOST, which nothing in the repo sets.
utils/diagnostics/container.py resolves without a marker: RESCUE_DEPTH is
a recursion base case, and the integer 0 says what the string "0" said.
The file sits exactly at the 500-line cap, so a comment line was not
available.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
get_merged_domains gated the onion injection on node_onion being set AND
svc-net-tor being in group_names. When the node was empty it answered the
clearnet map with no error and no log line, and because group_names is
part of the cache key, a clearnet map and an onion map coexist in one
process under separate keys. One render then serves onion while another
freezes a clearnet asset URL into a stylesheet.
That URL is unrecoverable at request time: body_filter.lua returns a CSS
response before onionize(), so the <link href> in the head is rewritten
while the url(...) inside the stylesheet body is not. The symptom is a
single toHaveCSS mismatch and nothing else.
An empty node under a deployed svc-net-tor now raises. A scope carrying no
group_names at all stays silent, because that is indistinguishable from a
legitimate non-tor deploy.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
test_container.py decorates with @unittest.mock.patch but imports only
unittest. A submodule is not bound as an attribute by importing its
package, so the name resolves only while some other module in the same
interpreter has already imported unittest.mock.
That makes the defect invisible in a full run, where another suite
imports it first, and fatal under a narrow INFINITO_TEST_PATTERN or on
an xdist worker that receives this file alone: collection dies with
AttributeError: module 'unittest' has no attribute 'mock'.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Resolved 35 conflicts per the designer x store-i18n checklist: store-i18n stays the single point of truth for credentials and role logic, designer's verified design work is kept.
Credential convergers: store-i18n's realign contract wins where both sides have one (mediawiki realign_admin_password.sh over converge_admin.php, pgadmin realign over administrator.yml, matomo utils/converge_superuser.yml over 09_admin_password.yml); the designer-only ones with no counterpart stay.
READMEs: test_readme.py hard-fails a committed Quick Setup or Credits block, both of which designer's side carries, so those go and only the Further Resources heading rename and the gallery links remain. SuiteCRM takes main's session fix, leaving exactly one SUITECRM_SESSION_CONFIG_HOST and dropping the Symfony cache-rebuild task the entrypoint now covers.
Two defects the rename detection hid: git paired store-i18n's web-app-meta against designer's web-app-sphinx, so sphinx's design slots would have been lost and its require(./test-design) landed in web-app-meta, whose spec has no such file. The sphinx design spec and its require now sit in web-app-docs, the role store-i18n renamed sphinx to.
Designer's eleven new Playwright design specs read env with || "" defaults, which test_unguarded_env_default.py refuses because an empty string matches every substring assertion; 35 bindings now go through requireDotenvValue.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
templates/style.css.j2 maps page, panes, object tree, tabs and primary
actions onto the design tokens and makes the menu bar the brand frame;
templates/design.js.j2 follows pgAdmin's native theme switch. The logo
and login-logo slots meta/services.yml declares sit on frame chrome, and
with design on templates/env.j2 sets PGADMIN_CONFIG_APP_NAME to the
configured title.
pgAdmin keeps its internal user in its own database, and
PGADMIN_DEFAULT_PASSWORD only seeds the first start. tasks/main.yml now
flushes the compose handlers and, on the stack host outside server
mode, runs tasks/administrator.yml with files/python/administrator.py:
it creates the internal administrator if missing, otherwise realigns
password, active and locked state, failed login attempts and the
Administrator role. The identity is the platform administrator as on
feature/store-i18n: PGADMIN_ADMIN_EMAIL and PGADMIN_ADMIN_PASSWORD in
vars/main.yml, meta/users.yml declares the user, and meta/secrets.yml,
whose only credential the role no longer reads, is gone.
tests/unit/python/roles/web-app-pgadmin/files/test_administrator.py
covers the script. meta/services.yml names the container pgadmin for
PGADMIN_CONTAINER_ADDRESS.
files/playwright/test-design.js checks tokens, that the administrator
signs in with that password and holds admin rights, panes, tree, tabs,
the menu bar frame and its focus indicator, primary actions, the theme
switch, logo, favicon and title, and captures the gallery against the
role's own database server; templates/playwright.env.j2 hands the
platform administrator, the design URLs and the database to it.
README.md links the gallery; TODO.md records the Gravatar request and
the memory pressure at the 256m limit.
The design was verified green on the compose stack (10 passed); the
switch to the platform administrator is checked statically only, the
first live run is the verification on the merged state.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Open WebUI loads /static/custom.css on every page, so tasks/00_core.yml
renders templates/custom.css.j2 when design is on and meta/volumes.yml
mounts it read-only there; it maps the sign-in page, the chat shell,
Markdown, tables and code onto the design tokens and leaves the OLED
dark theme to the user. templates/design.js.j2 mirrors the native theme
setting into data-design-theme. meta/services.yml keeps logo and title
replacement off, so Open WebUI keeps its own branding.
With sso and ldap both off, tasks/utils/admin.yml runs files/python/
administrator.py, which creates the native administrator or resets its
password and promotes it, so the native sign-in has an account; the
identity is lookup('users', 'platform_administrator') as on
feature/store-i18n, declared in meta/users.yml next to the user the
role still reads elsewhere. tests/unit/python/roles/web-app-openwebui/
files/test_administrator.py covers the script with fake values.
files/playwright/test-design.js checks tokens, sign-in page, chat shell,
a stored answer, the native theme setting, the provisioned
administrator, the in-house custom.css, logo and title, and captures
the gallery; files/playwright/_shared.js waits for editable password
fields. README.md links the gallery; TODO.md records the open
self-registration with sso and ldap off and the bot password and
bearer tokens in container exec argv.
The design was verified green on the compose stack (10 passed); the
switch to the platform administrator is checked statically only, the
first live run is the verification on the merged state.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Twelve files conflicted; most resolutions keep both sides rather than pick
one.
cli env.py and compose.py: both branches extend the same spots. The tools
override is appended before write_override, because that call filters the
mirror override to the compose files it is handed. compose.py keeps this
branch's CACHE_ROLE_SHELL_DIR, since the cache scripts moved into a role and
main's literal scripts/docker/cache path is their old home, and takes main's
mirrored init image alongside it.
utils/docker/image/discovery.py: this branch's version_variants expansion
plus main's whole runner-image block, which runner_mirror imports.
sys-svc-compose network create: main's side wins on substance and this
branch's on structure. main carries eda6249d7b, the fix for a probe that
matched an id prefix and let a sibling shadow the exact name; this branch had
refactored the task into create/overlay.yml and create/bridge.yml but brought
back the docker_network_info probe that fix replaced. The subfiles stay and
now probe through the exact-name lookup. main's half also read
SWARM_CLUSTER_AVAILABLE, which this branch removed in favour of
IS_SWARM_MODE, as the merge base confirms, so the gates follow this branch.
The lookup's import follows this branch's get_entity_name to entity_name
rename.
web-app-pixelfed: the custom entrypoints stay removed, since the new image
brings its own and the variables behind them no longer exist, and main's
host_alias=true comes along.
web-app-suitecrm: main's session.yaml and its config/packages path, because
only that version wires framework.session and only that directory is read by
Symfony.
web-app-odoo: main's onionOptions import is required by the test.use below
it; this branch's test-seaweedfs registration stays.
Four defects surfaced that no conflict marker shows.
The auto-merge of web-app-suitecrm/vars/main.yml duplicated
SUITECRM_SESSION_CONFIG_HOST, which ansible-lint rejects. Removing it exposed
the second: SUITECRM_SESSION_CONFIG_DOCKER still named config/session.yaml
while the mount had moved to config/packages, so the freshness comparison in
tasks/01_manager_ops.yml would have tested a path nothing writes.
svc-db-elasticsearch carried environment: on two include_tasks, which Ansible
accepts on include_role only. ELASTIC_PW now sits on the five tasks that read
it, which is also the tighter scope for a credential.
The openclaw templates render NETWORK_TRUSTED_PROXY_CIDRS while their test
renderer passes only its explicit kwargs, so seven tests raised Undefined.
That predates the merge: the template line is on this branch and not on main,
and both sides share the same test file.
test_runner_mirror neutralises four environment variables under clear=False
and missed three, so the gpu and tools overrides leaked in from the container
and the assertion on the layered list broke. It now neutralises all seven.
make quality-high passes both phases.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Listmonk serves its admin and public pages with the custom CSS and JS it
keeps in the settings table, so the role writes the design there:
tasks/03_design.yml, included after the settings reaper in
tasks/02_manager_ops.yml, pipes templates/design/appearance.sql.j2 into
psql. With design on it writes the marked admin stylesheet, admin script
and public stylesheet, but only over an empty value or one that starts
with the role's own marker, so custom CSS an operator wrote stays; with
design off it empties only its own values. The same SQL sets
app.logo_url, app.favicon_url and app.site_name from the design lookup
and restores the former values when design is off; those keys move from
LISTMONK_SETTINGS to LISTMONK_DESIGN_BRANDING in vars/main.yml.
templates/design/admin.css.j2 maps the colors of the Bulma and Buefy
admin onto the design tokens, templates/design/public.css.j2 does the
same for the subscription pages, and templates/design/admin.js.j2 keeps
favicon and title on the configured ones while the single-page app
rewrites its head. meta/services.yml declares the 182x34 logo slot of
the navbar; the emptied files/style.css goes.
files/playwright/test-design.js checks tokens, public pages, the
administration, navigation and tags, the code editors, logo, favicon and
title, that the role writes only its marked settings, and captures the
gallery; the native form takes ADMIN_NATIVE_PASSWORD. The unit test
renders the SQL for design on, design off and design off without a
fallback. README.md links the gallery; TODO.md records the settings task
that always reports changed, the secrets it renders into the shell argv,
and the logo URLs taken unguarded when the logo replacement is off.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
report_failures asked `gh api --paginate` for the run's jobs with the jq
filter `[.jobs[] | {name, conclusion}]`. gh applies that filter to each page
and prints one array per page, so a run with more jobs than one page holds
produced two JSON documents, and json.loads stopped with "Extra data: line 2
column 1". Main run 38218402254 (113 jobs) failed its "Report main role
failures" job that way, and no issue was opened for its failed roles.
The listing now emits one job object per line and parses the lines one by
one, the same way logs/download/github.py lists jobs.
Verified: the old parse of the real output of run 38218402254 raises the
same "Extra data: line 2 column 1 (char 8491)" as CI; list_jobs on that run
returns all 113 jobs, 16 of them failed, grouped into 12 failed roles.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Funkwhale stores its theme in global preferences, so
files/python/design_preferences.py sets the instance name, primary,
accent and background color, ui__custom_css from
templates/custom.css.j2 and the logo and banner as own uploads. It only
writes while ui__custom_css is empty or starts with the role's marker;
custom CSS an administrator wrote reports FOREIGN and stays untouched,
and design off takes back only the role's own values. tasks/main.yml
flushes the compose handlers and includes tasks/01_manager.yml, and
templates/compose.yml.j2 names the api container so container exec can
reach it.
tasks/01_manager.yml probes the administrator with
files/python/admin_state.py (MISSING, DRIFT or OK for active, staff,
superuser and password), creates it with fw users create --superuser
or realigns password and rights with fw users set, the password always
through the task environment. The account is platform_administrator:
meta/users.yml and the ADMIN_* lines of templates/playwright.env.j2 are
identical to feature/store-i18n (3c72c2002a), so the host
administrator's sudo password never becomes a Funkwhale password and
the later merge keeps these hunks clean. store-i18n provisions no
Funkwhale administrator at all.
templates/design.js.j2 follows Funkwhale's own theme switch, and
files/playwright/test-design.js checks the tokens in both modes, the
instance theming with palette, custom CSS and pod name, the start page
hero and sign-in card, the signed-in interface, a theme picked in
Funkwhale, logo, favicon and title, the platform administrator's sign-in
with administration rights and a visitor's refused sign-up, and captures
the gallery. Unit tests cover
both scripts. TODO.md records the Typesense key in the sidecar's argv
and the icons the front fetches from api.iconify.design.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
BookWyrm styles its pages from compiled Bulma themes, so the design
arrives as one of them. files/scss/infinito.scss maps the theme onto the
design tokens; the Dockerfile copies it into css/themes/ and
tasks/00_core.yml stages it in the build context and triggers the build.
The old files/style.css overrides are dropped, the theme replaces them.
tasks/01_site.yml reads the generated logos and runs
files/python/site_design.py through manage.py shell, with every value
handed over in the task environment. The script provisions the
administrator as superuser with the admin and moderator groups and
realigns its password, sets the site name and the navigation logo, touch
icon and favicon, registers the corporate theme and makes it the default
only on a fresh site, and closes the setup wizard. With design switched
off it removes only its own theme and logos. It prints BOOKWYRM_RESULT,
which changed_when and failed_when read.
templates/design.js.j2 mirrors the light or dark theme a user picks in
BookWyrm into data-design-theme, meta/csp.yml admits scripts, styles and
fonts from the object store that serves the static files, and
meta/services.yml declares a 100x100 navigation logo slot.
files/playwright/test-design.js checks the tokens in both modes, the
theme's page, primary action, selection and text, the divider, the
closed setup wizard, an upstream theme a user picks, title and logo, and
captures the gallery.
TODO.md records the registration keys BookWyrm 0.9.3 never reads, the
stale reason that blocks the administrator persona, the SeaweedFS test
that only signs in through Keycloak, and the missing static files of the
variant without an object store.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
templates/style.css.j2 maps the colors of Baserow's compiled stylesheet
onto the design tokens, with the vendor pseudo-element lists split so no
browser drops a rule. templates/design.js.j2 sets the favicon, and
meta/services.yml declares a 160x32 logo and a 32x32 symbol slot.
files/playwright/test-design.js checks the tokens in both modes, the
sign-in page, the signed-in interface, the API documentation, logo,
title and the collapsed sidebar symbol, and captures the gallery.
Baserow's sign-in form asks for the e-mail address and authenticates it
against the username. files/python/bootstrap_admin.py therefore stores
the superuser under its normalized e-mail address. It moves a superuser
kept under the username, or found by its first name after the address
changed, onto the current address, gives an adopted account staff and
superuser rights, realigns the password through check_password,
completes the profile and closes the first-administrator signup page.
It prints BOOTSTRAP_STATE, which the task's changed_when reads.
Ported from feature/store-i18n as identical text, so the later merge
keeps these hunks clean: the platform_administrator account in
roles/user-administrator/meta/users.yml as store-i18n has it now
(username platformadmin, administrator role), roles/web-app-baserow/
meta/users.yml and the BASEROW_BOOTSTRAP_ADMIN_* switch (3c72c2002a and
its follow-ups), and the argv form of the migration, bootstrap and MCP
tasks, which hand only the secrets through -e NAME plus environment
(177255329f). The bootstrap task keeps its own changed_when, because
store-i18n's 'created' never matches this script's output; the
password realign of 23e642a3f3 is part of the script.
templates/playwright.env.j2 reads ADMIN_* from platform_administrator,
the README persona note names the e-mail identifier, and TODO.md is
gone with the argv port that resolves it.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
After a version build the builder queued every translated language, not the
configured ones. The list now reaches the library: server.py reads
DOCS_I18N_PRELOAD_LANGUAGES, Library freezes it into a set that drops the empty
string ''.split(',') yields when nothing is configured, and builder.py queues a
language only when it is in that set. Every other translation is built when a
visitor opens it.
The warm-up task accepted 202, which means queued, so the deploy walked on
while the sites were still building. A second task now waits for 200. Both read
DOCS_I18N_PRELOAD_PATHS, hoisted out of the loop it used to be written inline
in, because two tasks need the same language-code-to-URL-path mapping.
DOCS_BUILD_PARALLEL divides by overhead + 2 * process_mb instead of
overhead + process_mb: with a translation building beside the version build a
parallel slot holds two process-sized allocations. The new divisor is strictly
larger, so the value can only fall, never rise; with the configured 3072 and
2048 MB it differs from the old one only between 10 and 16 GB of service
memory, where it goes from 2 to 1.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
roles/web-app-n8n/meta/services.yml conflicted on the n8n pin: main already
carries 2.43.4 from #900, newer than this branch's 2.43.2, so main's side is
kept. The provisioning fixes of this branch are unchanged.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The repository is bind-mounted into the infinito container, and every deploy
runs `make setup` there, which depends on `dotenv`. The generator therefore
rewrote the host's .env with the values the container sees: the host lost
INFINITO_RUNNING_ON_GITHUB=true that way, and the guide replay pulled
alpine:latest past the GHCR mirror.
compose.yml now marks the infinito service with INFINITO_DOTENV_HOST_OWNED,
and `python -m cli.meta.env` keeps an existing .env while that marker is
"true". A missing .env is still generated, and `make dotenv-force` (env -i)
still rewrites it. The key stays out of the env registry: only compose.yml
sets it, it is no configuration value, and a .env entry could freeze the
host .env.
test_distros now also strips INFINITO_ENV_LOADED. load.sh returns early
while that guard is set and the mtime of .env is unchanged, so the test only
reloaded the budget keys it strips because the container kept rewriting .env.
Verified: unit tests cover a kept .env, a missing one being generated, and a
rewrite without the marker. Local compose deploys print "Kept .env: the host
checkout owns it" where they printed "Wrote .env" before, the host .env
checksum stayed the same across a deploy, and the four keys holding host
paths resolve inside the container under the same paths.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The network probe used community.docker.docker_network_info, whose
get_network() scans GET /networks and returns the first entry whose Name
equals the requested name OR whose Id starts with it, in a list order that
changes between calls. In main run 38038541631 (Zammad#1, swarm) the
central redis overlay got the ID n8nwzb1q39m3u4k7awwwmt6yg, the probe for
the not yet created provider network "n8n" returned it as existing, the
create step was skipped, docker stack deploy attached hermes to the redis
overlay as "n8n", and the oauth2-proxy then reached the password-protected
central redis by its bare host name: "NOAUTH Authentication required".
The same scan can also return the ID-prefix twin of a network that does
exist, so comparing the returned name alone would turn that network into
"missing" and fail the create with "already exists".
Both probes now run `container network ls --filter name=<entity> --format
'{{json .}}'`, whose name filter never matches an ID. The new lookup
network_driver picks the entry named exactly after the entity and returns
its driver, '' when there is none; the drift checks, their messages and the
create steps all read that one entry. No docker module is left in the file,
so its two docker-module-uses-container-cli suppressions go too.
Verified: on the host daemon GET /networks/<id prefix> returns a network of
another name and docker network ls --filter name=<id prefix> returns
nothing. In ansible-playbook against real docker network ls output the
lookup returned the exact network's driver in 20 of 20 probes beside an
ID-prefix twin, an empty driver (create path) for a missing network and for
a name only contained in a sibling, and the drift path for a foreign
driver. Unit tests cover the same cases; with a substring match instead of
the exact name the two sibling cases fail.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
cells() awarded the priority glyph only in the branch that has a chunk, so a
priority row beyond the budget rendered as a bare disabled mark,
indistinguishable from an ordinary row the sweep did not reach. The two need
different answers: --offset indexes the regular line only, so it never reaches a
missed priority row, which needs a shorter priority line or a run of its own.
The status now carries both glyphs and the module docstring says what to do
about it.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
A lockup slot on a page or panel drew its title in the light
--design-primary with a --design-on-primary outline in both color schemes.
With an orange base that is about 3:1 on the light surface, and the SVG
never changed in dark mode; the Sphinx, Jenkins and Jellyfin passes
reported it.
sys-front-inj-design now passes --design-link with a --design-surface-1
outline for light and their dark values for dark. render_assets() hands
the dark pair to every slot without frame: true, and render_svg() adds a
style rule under prefers-color-scheme: dark that overrides the fill and
stroke attributes. Slots on brand chrome keep the frame colors, and the PNG
keeps the light pair. The design_assets action requires the two new
arguments; its two callers, assets.yml and Keycloak's theme task, pass
them.
A unit test asserts the dark rule on page slots and its absence on frame
and logo-only slots; it fails when the dark pair is not handed on. In
headless Chromium an <img> of the SVG renders the title #825200 on #fff5ea
in light and #ffa500 on #140e07 in dark.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The design queue skipped a hand-kept UI_LESS_ROLES list in
cli/meta/roles/design. The decision now lives in each role: a web-app- or
web-svc- role without a UI to design sets design.enabled: false in its
meta/services.yml, and design_disabled() reads that flag. web-svc-design,
the provider, stays out by its entity name.
A new lint, test_design_off_reason.py, requires a design entry in every
web-app- and web-svc- role and, where the entry is switched off, a
"# nocheck: design-off Reason: <text>" on the design: line or the comment
lines directly above it. The rule is in the suppression catalog, and
requirement 037 decision 13 describes it.
Twelve roles switch design off with a reason checked against their code:
asset and simpleicons were already off; file, onlyoffice, bridgy-fed and
mig were on and no longer receive the shared CSS and design script; cdn,
collabora, coturn, legal, mirror and xmpp had no entry. The four that were
on drop their bond and their design variant overrides, which the
disabled-bond and variants-static-override lints reject on a literal false.
The queue lists 77 candidates; HEAD's list also counted bridgy-fed, mig,
collabora and onlyoffice. Cosmos diagrams and the integration matrix are
regenerated.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Keycloak stores the permanent admin's password in its own database. When the
inventory value rotates, the next deploy renders a credential that
authenticates against nothing, and the role cannot reach the instance it just
restarted. The bootstrap account is no way back either: Keycloak writes
KC_BOOTSTRAP_ADMIN_* once, on an empty database, and upstream states it is
meant to be removed once a permanent admin exists.
The role now keeps the credential its predecessor used. Before sys-stk-full
re-renders the environment file with force: true, utils/capture_previous_admin.yml
reads KEYCLOAK_PERMANENT_ADMIN_PASSWORD out of the copy still on disk. When the
login after the restart is refused, utils/converge_previous_admin.yml signs in
with that captured value and moves the account onto the newly rendered one,
which the restarted container already carries in its environment. The bootstrap
fallback stays what its name says: the path for an instance that has never been
deployed. A host with no previous file captures nothing and goes straight there.
Reading the file back needs the inverse of dotenv_quote, which did not exist.
dotenv_unquote switches on the same DEPLOYMENT_MODE as the forward direction, so
a value written in compose is unescaped and a value written in swarm is taken
verbatim; dotenv_value pulls one assignment's raw right-hand side out of the
rendered text. The password never reaches argv: the convergence login passes it
through container exec -e plus the task's environment:.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Provisioning merges into host_vars and never deletes, so the false
enabled/shared toggles that apply_services_disabled writes for a key stayed
in the round inventory after a later run dropped that key from disable=.
That covered the provider's own entry too: with seaweedfs once disabled,
web-app-seaweedfs kept shared: false, the service registry offered no
seaweedfs provider, container_networks attached no consumer, and BookWyrm
ran USE_S3=true against http://seaweedfs:8333 without a route.
reconcile_services_disabled_from_env records the keys of each disable run
in the host_vars key _INFINITO_SERVICES_DISABLED. On the next run it deletes,
in every application block, the enabled/shared toggles that are still
exactly false for a recorded key the current disable no longer lists, drops
a block left empty, and rewrites or removes the record. It runs right after
ensure_host_vars_file, before credentials and the variant bake, so a variant
that wants a toggle off writes it again. With an unchanged or empty disable
it does nothing. An inventory written before this change has no record, so
its older toggles are only reverted after one run with the old list.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The GHCR mirror covered role images only. The images the runner pulls on
its own went straight to Docker Hub: coredns from compose.yml, the cache
stack in compose/cache.override.yml and alpine:latest, the init image of
scripts/docker/cache/package-frontend/certs.sh. With the anonymous
githubactions quota spent, the guide step of Postmarks#0 (main run
37954203718) and Shopware#2 (#872) died on "toomanyrequests" while pulling
alpine:latest, after coredns had hit the same limit.
iter_runner_images (utils/docker/image/discovery.py) finds those images,
and the mirror sync and the fork-PR wait now cover them next to the role
images. On a real GitHub runner compose_file_args layers a generated
build/compose/runner-mirror.override.yml that points only the services of
the layered compose files at their mirror, and the certificate generator
gets the mirrored init image. Local runs and act are unchanged, and the
coredns literal stays in compose.yml for Dependabot. test-dns now waits for
images-mirror-missing like the deploy and workspace jobs, since it brings up
the stack too.
Verified with docker compose config on a simulated GitHub runner: the
service set is unchanged and only the runner images point at
ghcr.io/infinito-nexus/core/mirror/docker.io/..., and an existing role mirror
under that scheme pulls anonymously.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Twelve jobs of run 37865281419 failed on the docs language spec:
tests/test-languages.js:25:7 the deployed working tree is published in de
with a language switcher (54.6m)
Error: Expected the de site of the deployed working tree to be built
The spec polls /deployed/de/ until the suite budget runs out, so a build that
hangs is reported as a spec timeout with nothing to read. None of the five
subprocess calls the build makes carried a timeout, so any of them could hold
the queue open forever:
library.py _git() rev-parse, tag --list and remote update --prune
library.py fetch() git clone --mirror
i18n.py each generator of generate_commands
i18n.py sphinx -b gettext
builder.py _run() the sphinx build of one version
The two git paths reach the network, which is where an unbounded call hangs
without failing.
`_run` streams its child's output into the build state, so it cannot take
`communicate(timeout=)` without losing the progress that `_append_log` and
`_save_state` record, and Popen has no timeout of its own. A threading.Timer
kills the child instead. The kill lands in the existing returncode branch, so
the build is reported through `_failed`, which already writes the log tail into
the state file.
RUN_TIMEOUT_SECONDS is set to catch a hang, not to beat the spec's clock: no
measurement of a successful de build exists, so a tighter value risks turning a
slow build into a failing one. With one, the value belongs below the poll
window the spec scales to.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
`disable=lam` aborted the local deploy of web-app-openbao with "`disable`
conflicts with the current inventory state", and web-app-moodle carries the same
construction in two of its variants.
The writer and the conflict check disagreed on scope. apply_services_disabled
walked `find_roles_with_service`, the roles whose meta/services.yml declares the
key, so neither role was in that set for `lam` and nothing was written.
find_services_disabled_conflicts walks every deployed app's host_vars instead,
found the `lam` block the variant had baked, and refused the run. Dropping
`lam` from the list is not an answer either: the round then deploys web-app-lam
although ldap is off.
The writer now also flips a key it finds in host_vars under an app whose role
does not declare it, and only the flags that are already in that block. It never
creates a key or a flag, which is what `_declared_toggles` exists to prevent:
compare_application_keys rejects a host_vars key the role contract does not
carry. Overwriting a value the variant already wrote introduces nothing, so that
rule is untouched; test_skips_app_whose_role_does_not_define_service still
passes and pins it.
The alternative the TODO proposed, declaring `lam` in meta/services.yml with the
`'web-app-lam' in group_names` flags, is the worse trade: a consumer entry is
resolved through the service registry, so it would bind web-app-lam and its
whole closure into every round that carries openbao or moodle, for a key both
roles read only to gate a Playwright spec.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
A role that exports an outbound proxy for one purpose applies it to every
HTTP call the app makes, so an internal dependency reached by name is proxied
too. web-app-pixelfed exported the Tor SOCKS proxy container-wide for its
onion OIDC issuer and its S3 client then failed with
cURL error 97: cannot complete SOCKS5 connection to seaweedfs
for six minutes per attempt, until sixty retries had consumed the whole
suite budget and the job was killed. web-app-espocrm ships a byte-identical
proxy block and configures its S3 endpoint from the same lookup.
web-app-nextcloud sets the proxy globally in proxy.config.php and listed two
internal hosts in proxyexclude, neither of them the object store it reaches
by hostname.
9d91fbea83 had tried to retire the per-app proxies in favour of onion routing
over DNS, and 20ddda8c1c restored them for the OIDC issuer, so a broad proxy
is the current design and the exemption list is load-bearing.
lookup('proxy_excluded_hosts', application_id) is now that list: the loopback
forms plus the object store host when the role stores objects. It is a lookup
and not a group_vars value because no group_vars file resolves
application_id, so a global would have depended on which role happened to be
running. Nextcloud keeps its hand-written list of public hosts; only the
internal half is derived.
The lint gained a second rule for the class rather than the symptom: the
nextcloud list named no loopback literal at all, so the literal rule could
never have fired on it. A variable whose name marks it a proxy exclusion list
must render the lookup, unless it only composes other exclusion variables, and
a ..._PUBLIC_... list is exempt because it names no internal service.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Both provisioners follow nextCursor for workflows and step skip for API
keys until a short page arrives. A cursor n8n repeats, or a key listing
that ignores skip, would keep that loop running forever and hang the
deploy instead of failing it.
workflows_named now remembers every cursor it followed and exits with
"FAILED listing workflows: nextCursor repeats" when one comes back.
api_keys_labelled exits when a non-empty page brings no key id it has not
already seen. Normal paging is unchanged.
Four regression tests drive both scripts against a fake API that repeats
its cursor or ignores skip; without the guards exactly those four fail.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
tests/lint/repository/structure/test_python_file_size.py caps a Python file at 500 lines and asks for a split rather than a raised limit, and engine.py reached 522 once the network claim took a lock. engine.py keeps the engine client, the labels, EngineError and the compose backend at 273 lines; swarm.py takes the swarm backend at 258.
The methods moved verbatim, so nothing changes behaviour. swarm.py imports the labels and the error from engine while engine imports nothing back, which is what the first attempt got wrong: a Pool mixin, following the build_queue.py precedent, would have made the two modules import each other. server.py splits its import, and agents.py reads only labels so it is untouched.
The pool tests follow the code into test_swarm.py, the mirrored path test_mirrored_unit_tests.py asks for, and test_engine.py keeps the transport, connect and compose cases.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
c4ea94c778 read the 404 from POST /networks/<id>/connect as the swarm allocator not having realized the overlay yet, and retried for realize_timeout. 03ed2f1ca1 then recorded the real constraint: a swarm service task cannot be attached to a network it was not created with, so the retry could only burn 131 to 133 seconds per attempt and backend.create was never reached (run 37252453141, job 111587113070).
ensure_pool therefore creates MAX_RUNNING overlays up front and adds the missing ones to its own service spec, which costs one reschedule of this task per deploy and none afterwards. network_for then hands an agent the lowest overlay nobody holds, and gives back the one it holds already; _scale writes that network in the same update that lifts the replica count, so claiming a slot adds no further restart. A pool network carries the broker and at most one agent, so the isolation the README promises holds.
The claim is taken under a lock and remembered until the agent stops, and a service counts as holding its networks from the moment the engine records its desired replica count rather than from the moment a task reports running. Without both, two owners starting their first agent at once are handed the same network and reach each other: four concurrent claims against the unguarded form all returned the first slot. Because the claim and the agent keys live in one process, ensure_pool refuses to start when the broker itself is scaled past one replica.
Compose keeps attaching at runtime, because a bridge network is local and realized at create. connect stays for that path, with its docstring corrected to name which mode calls it.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
7916c0177e wrote the cap as {{ IAM_USERNAME_MAX_LENGTH }} inside the {% set _base_attrs %} block. A set body is already a Jinja expression, so the nested delimiters are a syntax error rather than an interpolation, and jinja2 answers TemplateSyntaxError: expected token :, got }. realm.json.j2 includes the component unconditionally, so every realm render raised, not just the username path: that single typo is the whole of the 9 unit and 7 integration failures the gate reported.
The test context gains the value because render() builds its environment with StrictUndefined, which raises on an absent name instead of rendering empty, and a new case asserts the rendered validations.length.max equals it so the cap cannot fall back to the Keycloak default of 255 unnoticed.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
`make quality-high` rewrote these files on every branch: the SuiteCRM
README and the root role table lacked the Redis service #881 added, the
session path test was not formatted, and an OpenTalk secret description
from #873 was not wrapped.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
n8n 2 returns ten API keys and a hundred workflows per page by default. A
stale managed key behind the first page survived, and a managed workflow
behind it was created a second time. Both scripts now follow every page,
narrowed by label or name, and keep the exact match in the script.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Autoformat output on the test PR #881 added: the redis_session_path call
exceeded the line width, so its arguments move onto their own line. The
token sequence is unchanged, so the assertion still proves what it did
before, that a password carrying &, =, ?, # and a space survives the
connection URL intact instead of splitting it into extra parameters.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
n8n 2 pages GET /rest/api-keys as {data: {items, counts, totals, owners}},
ten keys at a time, and matches `label` as a substring, so walking the body
raised "'str' object has no attribute 'get'". It also enforces API key
scopes, so activating the MCP server workflow needs `workflow:activate`.
Both scripts now read every list and create response through one
normalisation that unwraps the page, look the managed key up by label, and
name an unexpected shape by its keys only. The normalisation is taken over
from a392e2dd49 on feature/store-i18n.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
