A role states which package it needs; the registry owns every distribution
mapping and every acquisition path. Declarations live in the repository-root
meta/packages.yml when more than one consumer installs them, and in a role's
own meta/packages.yml otherwise.
schema.py owns the declaration shape, registry.py finds and indexes the
declarations, plan.py turns a resolved spec into ordered module calls, and the
package_install action plugin only executes them. Keeping the plan pure makes
every acquisition path unit-testable without an Ansible connection:
distribution repositories, the AUR with its unprivileged build user, COPR, PPA,
an EPEL bootstrap package, the pacman multilib section and source builds.
The data is guarded by lints: a package id and a concrete distribution package
name each belong to exactly one declaration, a role installs only its own or
shared ids, every id resolves for all five default distributions with an
empty mapping justified by a nocheck, declarations read in ascending order,
every plan builds and names only whitelisted modules, and an external test
queries the real package indexes of all five distributions.
This commit adds the mechanism and the shared collection; no consumer uses it
yet.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The drill configured svc-bkp-remote-2-local with a fixed manager plus NFS provider list for every app, while the manager only runs a backup repository when the app's include closure pulls svc-bkp-volume-2-local or svc-bkp-secrets-2-local in. For an app whose container_backup consumer is statically off, such as svc-dns-unbound, user-backup never ran on the manager, so the pull was rejected with 'Permission denied (publickey,password)' and the fail-if-any wrapper took the whole unit down at drill step 4/9. Four roles reach that step today: svc-dns-unbound, svc-db-redis, svc-registry-cache and svc-registry-docker.
backup_repos.py expresses the placement once and matrix derives the provider list from it per round. derive_includes accepts the round's variant map because INFINITO_APP_VARIANTS only reaches the subprocess callers. The extras file keeps the device paths the drill reads back and loses the provider list nothing read from it, and the drill banners stop claiming backup legs that were never installed.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The drill configured svc-bkp-remote-2-local with a fixed manager plus NFS provider list for every app, while the manager only runs a backup repository when the app's include closure pulls svc-bkp-volume-2-local or svc-bkp-secrets-2-local in. For an app whose container_backup consumer is statically off, such as svc-dns-unbound, user-backup never ran on the manager, so the pull was rejected with 'Permission denied (publickey,password)' and the fail-if-any wrapper took the whole unit down at drill step 4/9. Four roles reach that step today: svc-dns-unbound, svc-db-redis, svc-registry-cache and svc-registry-docker.
backup_repos.py expresses the placement once and matrix derives the provider list from it per round. derive_includes accepts the round's variant map because INFINITO_APP_VARIANTS only reaches the subprocess callers. The extras file keeps the device paths the drill reads back and loses the provider list nothing read from it, and the drill banners stop claiming backup legs that were never installed.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
A sandbox node runs gVisor as its daemon-wide default runtime and every
container scheduled there inherits it, so keycloak failed to start on one.
Swarm has no taints, which leaves the inverse placement constraint as the only
lever: sandbox consumers are pinned to the labelled nodes, everything else is
kept off them. Both halves were measured on a live cluster first, because a
docker constraint has no presence test and rejects an empty comparison value:
`!= true` does match a node without the label, and three replicas under it
avoided two labelled workers entirely.
That makes the set of labelled nodes load-bearing for the whole platform, and
it was decided by accident. Without an `svc-virt-kata` inventory group the
service loader runs the role on every host, so every worker becomes a sandbox
node and the rest of the platform is evacuated onto the stack host. The role
now refuses to run in swarm mode until the operator names those nodes.
With the group present the opposite gap opened: the provisioner puts every role
group on the single `--host` it is given, so the role lands on the manager
alone, the label task skips itself there, and no node ends up labelled. The old
assert only proved that a worker exists, which says nothing about where the
role ran, so the deploy passed the check and then hung until the convergence
wait gave up. The assert now reads the labels back from the stack host.
`docker node ls --filter label=` matches engine labels, not swarm node labels;
the filter has to be `node.label`.
The constraint itself is emitted only where a sandbox tier exists, so a deploy
without an agent role is untouched by any of this, and the lab dedicates a
single worker: labelling both would leave the manager as the only home for
every other service.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
`svc-virt-kata` is pulled in as a shared service of the agent roles, so it also
executes on hosts that are in no inventory group of that name. Gating the
runtime registration on `'svc-virt-kata' in group_names` therefore left such a
host with runsc installed and its node labelled kata-capable, while
`daemon.json` registered nothing and the daemon kept running everything under
runc: the exact isolation breach the label promises is closed. The stat probes
in `daemon_config.yml` decide instead, and they cannot be fooled by an
inventory that says otherwise.
Nothing else in the tree installs or references runsc, so a shim on disk is
proof that this role put it there.
The agent compose templates pin their runtime on the same `services.kata`
switch that already drives the swarm placement constraint, instead of pinning
unconditionally against a daemon config written under a different condition.
Both roles declare that key themselves, so the lookup is the strict form: a
deleted flag now aborts the render rather than silently starting the agent
without a sandbox.
The new unit test covers what nothing covered before, which is how the runc
breach went unnoticed: an absent shim is never registered, a default runtime is
always a registered one, the stack host keeps the shared kernel, compose never
sets a daemon-wide default, and registration follows the probe rather than the
inventory group.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
guide_select built its random pool from invokable + lifecycle only, so
the Guide job could pick roles whose meta/tests.yml declares every
deploy mode untestable (svc-storage-nfs-client, which hard-asserts a
reachable NFS server). Filter the pool by the role's declared skip for
its guide mode; a malformed tests.yml now fails selection loudly
instead of being silently ignored.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
A role with a Model Context Protocol surface declares one mcp: entry in
meta/services.yml. The vocabulary module owns every allowed value, a hard
lint validates each declaration, and roles_with_service gains an opt-in
direction= kwarg so a client role resolves its servers at deploy time.
The lookup stays backwards compatible: without direction= the entries
keep their original four keys, so the existing dashboard, prometheus and
matomo callers are untouched. With it, each entry additionally carries
transport, auth, auth_subject and an endpoint dict whose port is
resolved from the referenced service's ports.local or ports.internal.
The lint rejects Jinja on MCP-specific fields (only enabled/shared may
carry it) because the static resolver cannot evaluate a templated block,
and it verifies that endpoint.service_key names a service in the same
file whose ports.* actually contain endpoint.port_key.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
LiteLLM rejects user-supplied virtual keys that lack the sk- prefix
(HTTP 400 on POST /key/generate), so consumers of the shared gateway
need a credential that already carries it. Mirrors the existing
base64_prefixed_32 precedent: sk- plus 48 secure alphanumeric chars,
51 total, clearing the min_length: 50 validation the consumer schemas
declare.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The --failed selection treats anything not green (cancelled, timed-out, still-running) as a failure, so retriggering a partially-running or aborted run re-fires jobs that never actually failed. Add --failed --strict to select only roles with a hard failure conclusion, excluding cancelled/aborted and running. New aliases i8cifo (strict base, mirrors i8cifail) and i8pufo (sign-push then strict retrigger, mirrors i8pufail); i8pufail is unchanged. Verified against a live run: --failed matched 48 roles (8 failed + 40 still-running), --strict matched exactly the 8 hard failures.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The asset lookup hardcoded https for the internal CDN origin, but
.onion vhosts are plaintext with no 443 listener, and the dashboard
mirror-rewrite sent even onion-local asset URLs through the mirror
(404). The adapter never loaded, so the OIDC code was never exchanged
and Login stayed visible after the round-trip.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
disable= now drops providers left with no surviving parent, not only
the named provider roles. Compose (init) and swarm (derive_includes)
share one reachability cut: BFS from the surviving primaries over
dependency + service edges with disabled roles as cut-nodes. No-disable
deploys stay byte-identical.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The ruff toolchain was unpinned, so host venvs, the lint container and CI
installed different releases over time. That drift is why lint-python only
failed on CI: locally ruff 0.15.x preview-gated ISC004/FURB192 and (falsely)
fired S310 on urllib.request.Request, while CI's 0.16.0 has ISC004/FURB192
stable and drops the S310-on-Request false positive.
Pin ruff==0.16.0 in pyproject.toml's dev extras (Dependabot keeps it current)
and have utils.install.lint.ruff read that pin and reinstall whenever the
found version drifts, so every environment resolves the same ruff. Then fix
what 0.16.0 flags: parenthesise the implicit string concatenations (ISC004),
use min() over sorted()[0] (FURB192), drop the now-unused S310 noqa on the
Request() builders (RUF100), and delete the narration comments that the
comment lint surfaces in the touched files.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
generate_cache_override pinned deb.debian.org/archive.ubuntu.com/security.ubuntu.com
to INFINITO_CACHE_PACKAGE_FRONTEND_IP for every buildable service whenever that
env var was non-empty. The IP is a baked default.env value, so the hijack fired
even where no cache frontend exists: the guide job sources .env, replays a role's
compose build inside the DiD stack, and the moodle apt build hit 172.30.0.4 with
'No route to host' because no frontend is provisioned in CI.
Gate the override additionally on the frontend CA file existing non-empty, the
same signal ca.sh uses for an active cache profile and that registry_cache_active
produces. When the cache profile is inactive (CI/guide) the CA is absent, the
override is skipped, and the build resolves the mirrors via real DNS. compose.py
is copy-deployed and cannot import Profile, so a filesystem signal is required.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Formatter output from make quality: the git ls-tree argument list in
complexity.model and the xwiki job-name dict in the report_failures test
wrapped across lines. No semantic change.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
report_failures.decisive_excerpt and new_in_branch._declares_application_id
now read via utils.cache.files.read_text instead of Path.read_text, and the
new_in_branch test constructs its vars path from ROLE_FILE_VARS_MAIN. Clears
the uncached-read and role-file-spot lint gates; behaviour is unchanged
(same UTF-8 decode, same exceptions, same resolved path).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
On a main run, report-main-failures parses the failed deploy jobs, downloads
each role's rescue-diagnostics artifact, and opens a 'CI failure: <role>' issue
(label ci-failure) carrying the run link, the artifact names, and the decisive
rescue excerpt inline. An existing open issue for the role gets a comment
instead. Mode emojis are resolved through the symbol glossary, not hardcoded.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
A push resolves the application roles that exist under roles/ locally but
not at origin/main and passes them as the orchestrator priority line, so
brand-new roles deploy and gate ahead of the regular matrix. The resolver
is stdlib-only (git ls-tree plus an application_id line scan) so it runs
on a bare runner without installing the resolver stack, and yields an
empty set when the baseline ref cannot be resolved.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Add an in_main column (🌳) to the role complexity matrix: it resolves
the role directories present under roles/ at origin/main once and flags
every row true/false, so a role that is new on the current branch stands
out. When the ref cannot be resolved (a non-git checkout or an unfetched
main) the marker falls back to present for every role rather than
falsely reporting them new.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Add tests/lint/repository/test_task_file_numbering.py: numbered task
files use exactly two leading digits, a numbered *_core.yml must be
00_core.yml, per directory the numbers are unique/consecutive and start
at 01 (00 with a core file), numbered includes appear in ascending
order, and shared helpers live in tasks/utils/ instead of the numbered
chain.
Apply it across every role: rename NN_core.yml to 00_core.yml with a
consecutive down-shift of the siblings, move letter-suffix children into
a subdirectory named after their parent task, relocate helpers included
from more than one place into tasks/utils/, and update every reference
(include_tasks, path_absolute, .ansible-lint-ignore, requirement docs).
Legacy narration comments in the touched files are deleted; the upstream
xwiki.properties reference block opts out via nocheck. The unused
MAILU_IP6_PUBLIC var (only ever named in now-removed comments) is dropped.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
pacman's XferCommand curl gains --retry-all-errors --retry-delay 2 so
connection resets during package downloads retry instead of failing the
update; plain --retry does not cover curl exit 35. No --max-time here:
package downloads are size-unbounded and a ceiling would truncate large
transfers. dnf gets retries=5 via its native config, matching the apt
Acquire::Retries hardening already in place for Debian.
Settle the egress lint follow-ups: collapse the nested retry checks per
ruff SIM102 and reformat the curl filter tests.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Add lookup('asset', app, pkg, path) and lookup('asset_host'): both resolve
the asset origin through one shared resolve_host, returning the internal
CDN domain when web-svc-cdn is deployed with flavor internal and
cdn.jsdelivr.net otherwise, so template URLs and CSP origins can never
diverge. Versions come from the role's committed files/package-lock.json.
Roles declare frontend dependencies in files/package.json; web-svc-cdn
aggregates every role lockfile, runs one npm ci per role in a node
container and mirrors the packages under /npm/<pkg>@<version>/ following
jsdelivr's path scheme, so internal and external URLs differ only by host.
Convert web-app-dashboard (keycloak-js 24.0.5 UMD) and web-app-littlejs
(bootstrap, fontawesome) to the asset lookup, drop the vendored keycloak
blob, and add a repository lint forbidding hardcoded external CDN hosts
in templates and CSP metas; pre-existing external CDN entries are checked
out with nocheck markers and a migration TODO.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Transient CDN/mirror TLS resets are a recurring hard-fail class (docker.com, PyPI). Harden all three egress paths at their SPOT: the curl filter now emits --retry-all-errors (retries=0 opts a call out), so every ansible curl retries; a new lint requires --retry-all-errors on file-downloading curl and --tries on wget in verbatim files/*.sh (which cannot use the filter), fixing the moodle/alias downloads it flagged; and the update role configures native retries for every package manager - apt Acquire::Retries, dnf retries=, pacman XferCommand over retrying curl - covering all five CI distros (ubuntu shares apt, centos shares dnf).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Conflicts: keep both new passthrough env vars; adopt the swarm-nfs
ollama manager-ops split while porting our model-pull retry hardening
into it. Strip three narration comments the stricter comment lint
rejects.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
When a run deploys both single-node modes, the host discovery redeployed roles the compose line already covers (e.g. web-svc-html in both matrices). The single-node workflow now signals a co-running compose line via INFINITO_HOST_EXCLUDE_COMPOSE (a run-context registry key like INFINITO_DEPLOY_MODE), and the host query excludes the compose line's actually-triggered selection before its own budget cut, so freed host slots go to roles compose does not cover. Priority and regular lines each filter against their own compose counterpart through the identical whitelist/blacklist pass-through; host-only runs stay unchanged.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Both the priority and the regular deploy line received the full per-mode slot budget, so a run with a priority input budgeted up to twice the mode share against the 256-job run limit. max_jobs now subtracts the jobs the priority line actually occupies - the regular line's blacklist is the priority whitelist, and the inner discover runs under the identical cap, so the deduction mirrors the real consumption including the priority line's own cut. A fully consumed budget short-circuits discover to an empty selection without invoking the complexity CLI (which rejects --max-jobs < 1), and the plan summary renders the reduced regular budget.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Resolves the runner Dockerfile to the lint-standard Acquire::Retries=3
and keeps the collabora TODO wording; hardens the merged tree for the
stricter lints (apt retries in the tor Dockerfile, shell-or-true
markers, comment cleanup).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
meta/services.yml carried a '# nocheck: single run-once seeder' marker whose token did not match the single-replica rule, so the seeder replicas:1 pin read as unjustified. Correct the token to single-replica (reason unchanged). test_plan.py is an autoformat reflow.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The repair services (rpr-container-soft/hard) emit a syslog 'infinito-kill' marker before they down/restart containers; the rescue collector captures containers/_kill-markers.txt (plus a 6h docker/containerd daemon journal) so a failure artifact names which service stopped containers.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Move the plan table out of the standalone orchestrator Plan job into each
deploy mode's discover job (swarm/compose/host), so the table shows as a
run-summary spot on the orchestrator that will run it. Drop the redundant
Priority column since the Triggered column already carries the star.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The ci trigger (--failed) and status commands matched deploy jobs only by
the compose/swarm glyph, so host-mode jobs (svc-storage-nfs-server etc.)
were invisible and re-triggering a failed run never picked them up. Add
host to the mode set and read the mode glyphs from the symbol glossary
(single source of truth) instead of hardcoding them. Align the host deploy
job titles to the glossary glyph 💻 (were a rogue 🖥️ absent from the
glossary) so they match, with a leading space like compose/swarm.
Priority (⭐) and regular (🔁) jobs were already both discovered (the match
ignores the line prefix); host now works in both. Past runs titled 🖥️ will
not match; a fresh run does.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Back to the pre-swarm-nfs-fix baseline:
- 5c2c8f531 + 985ec7eab: the swarm writer-quiesce DR drill. Deep-dive
attribution of CI run 29715516546 found it the dominant failure source
(17 of 23 jobs): [9b/9] handed-off generation mismatch, [6/9] quiesce
stack-rm network race, [4/9] remote-2-local pull stall, compose discourse
restore duplicate key.
- d6f980574: openproject swarm ingress to web:8080. Carried a leaking
proxy_internal_port lookup (no per-app default) that broke the seaweedfs
dependency front-proxy ('Key web not found'); the bypass-the-Apache-proxy
approach was never verified against downloads/X-Sendfile.
- 1dd153bad: xwiki replicas:1 + mem 3g. Did not resolve the xwiki superadmin
failure (still red in run 29715516546), so it is unverified.
Re-approach each from a clean baseline.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Follow-up corrections to b449f8157 (swarm DR drill):
- export test.env in 06/07 via set -a so the child db restore/probe
scripts inherit BKP_TEST_DATABASES_CSV and BKP_TEST_RESTORE_BIN
instead of aborting at their :? guards before any restore runs
- move db_probe.sh/db_restore.sh into a db/ subfolder (probe.sh,
restore.sh) and update all callers, so they stop tripping the
prefix-grouped-files structure lint
- define TEST_E2E_SWARM_DRILL in group_vars/all and drop the now
redundant default() from test.env.j2 (satisfies the undefined-var
and spot-default lints together)
- fix a mermaid parse error in the drill README: a ; in a
sequenceDiagram note was read as a statement separator
make quality-high green (test + lint phases, 15 targets, 0 failures).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Defer Swarm SQL replay to the disposable DR drill so shared writers such as Discourse are stopped before the destructive restore. Verify the selected generation with before/after database probes while preserving the Compose restore path.
Autoformat collapses now-fitting multi-line calls in the runtime model,
render helper and their unit tests. Format-only, AST-identical.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
parallel.sh mirrors its per-target wall-clock table into
GITHUB_STEP_SUMMARY, so make lint and make test render a target/status/
duration overview on the run page. The pytest runner writes junit XML
per suite into build/test-reports and test.yml renders them via
utils/github/pytest_summary.py: headline counts with total time,
failures opened with durations, every test collapsed slowest-first.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The swarm deploy job now renders the same summary page as compose:
role runtimes, a swarm structure section (topology roles plus live node
list and service placement from the manager) and the Playwright table.
The matrix deploy tees its ansible output and prints the compose
matrix-deploy round/PASS markers, so the runtime parser segments swarm
runs per variant round too.
The runtime table unifies tasks and roles with a Type column (tasks
first), appends the recap total row and flags a mismatch between the
parsed sum and the recap total; rows are read from ROLES RECAP sections
only, which removes the double-counted TASKS RECAP task rows. Each role
row carries one column per host: executed, skipped or failed, with
ignored failures downgraded. The Playwright variant column drops the
redundant variant- prefix. Verified against the logs of runs
29656268771 (swarm) and 29620158441 (compose): every segment's parsed
sum matches the recap total within 0.05s.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
