Extends the existing upgrade PR (branch upgrade-ce5308c) with the drive quartet v0.22.0 → v0.24.0 and nginx 1.31.6 → 1.31.7, on top of the changes already carried
here (redis 8.10.2, collabora 26.04.4.2.1, minio_initialize config v2). Rebased onto the
current upstream main (863691f): the earlier branch predated upstream's renovate.json, MAINTENANCE.md, and maintainer-metadata changes, which are now preserved (not deleted).
Image-tag diff vs upstream main
service
image
main (current)
this PR
app
lasuite/drive-frontend
v0.22.0
v0.24.0
backend / celery / celery-beat
lasuite/drive-backend
v0.22.0
v0.24.0
redis
redis
8.10.1
8.10.2
web
nginx
1.31.6
1.31.7
collabora
collabora/code
25.04.10.3.1
26.04.4.2.1 (carried from this PR)
config
minio_initialize
v1
v2 (carried; required for the Swarm config rename)
drive v0.24.0 — no breaking changes and no manual migration step: the DB is migrated
automatically on backend startup (verified live: all core.*, easy_thumbnails.*, malware_detection.*, rest_framework_api_key.*, sessions.*, sites.* migrations applied, migrate: done). Notable upstream changes are additive (rate-limiting on item creation,
storage-quota admin, WOPI sandboxed downloads, item-list speedups). Upstream's own dev compose
switched its local object storage from MinIO to RustFS — that is upstream's compose, not this
recipe, which continues to run the pinned MinIO sidecar.
collabora 26.04.x — the 26.04 image executes coolwsd directly (no shell in the image):
the old entrypoint/extra_params wiring is replaced by command args and a coolwsd --probe
healthcheck; the admin-panel password moves from the retired Docker secret collabora_p to the COLLABORA_ADMIN_PASSWORD env var (empty/unset ⇒ panel stays locked). Existing deployments are
unaffected by the orphaned collabora_p secret.
minio_initialize v1 → v2 — Swarm configs are immutable; the config content changed in an
earlier upstream change but the vendored version was left at v1, so upgrades of existing
deployments aborted with only updates to Labels are allowed. Bumping to v2 makes Swarm create
a fresh config object.
Advisory scan (step 2a)
Deterministic scan for v0.22.0→v0.24.0 (+ redis/nginx/collabora) reports 5 CVEs fixed, all
nginx (the five above). ⚠️ The scan's GitHub Security Advisories source returned HTTP 403
(rate-limited from the scan host) for every repo during this run, so that count is a floor — an
incomplete scan, not a clean one. redis 8.10.2's release notes carry security fixes without CVE ids;
collabora 26.04.4.2.1 (carried from the earlier run of this PR) resolves five GHSA advisories fixed
in the 25.04.11+/26.04.2+ lines.
Release command (operator)
The recipe version label (0.13.2+v0.22.0) is intentionally not bumped here. After merge,
publish with:
abra recipe release lasuite-drive -y
Not merged — awaiting operator review.
Tested green on the cc-ci recipe CI server (full suite, cold, against this PR head). NOT merged — for operator review.
## Upgrade lasuite-drive to drive v0.24.0
Extends the existing upgrade PR (branch `upgrade-ce5308c`) with the drive quartet
`v0.22.0 → v0.24.0` and nginx `1.31.6 → 1.31.7`, on top of the changes already carried
here (redis 8.10.2, collabora 26.04.4.2.1, `minio_initialize` config v2). Rebased onto the
current upstream `main` (`863691f`): the earlier branch predated upstream's `renovate.json`,
`MAINTENANCE.md`, and maintainer-metadata changes, which are now preserved (not deleted).
### Image-tag diff vs upstream `main`
| service | image | main (current) | this PR |
|---------|-------|----------------|---------|
| app | `lasuite/drive-frontend` | `v0.22.0` | **`v0.24.0`** |
| backend / celery / celery-beat | `lasuite/drive-backend` | `v0.22.0` | **`v0.24.0`** |
| redis | `redis` | `8.10.1` | **`8.10.2`** |
| web | `nginx` | `1.31.6` | **`1.31.7`** |
| collabora | `collabora/code` | `25.04.10.3.1` | **`26.04.4.2.1`** (carried from this PR) |
| config | `minio_initialize` | `v1` | **`v2`** (carried; required for the Swarm config rename) |
Upstream release notes (one per changed service):
- **app/backend/celery/celery-beat `v0.22.0→v0.24.0`**: https://github.com/suitenumerique/drive/releases/tag/v0.24.0 (and https://github.com/suitenumerique/drive/releases/tag/v0.23.0)
- **redis `8.10.1→8.10.2`**: https://github.com/redis/redis/releases/tag/8.10.2
- **nginx `1.31.6→1.31.7`**: https://nginx.org/en/CHANGES
- **collabora `25.04.10.3.1→26.04.4.2.1`**: https://www.collaboraonline.com/collabora-online-26-04-release-notes/
### Operator action required
- **drive v0.24.0** — no breaking changes and no manual migration step: the DB is migrated
automatically on backend startup (verified live: all `core.*`, `easy_thumbnails.*`,
`malware_detection.*`, `rest_framework_api_key.*`, `sessions.*`, `sites.*` migrations applied,
`migrate: done`). Notable upstream changes are additive (rate-limiting on item creation,
storage-quota admin, WOPI sandboxed downloads, item-list speedups). Upstream's own dev compose
switched its *local* object storage from MinIO to RustFS — that is upstream's compose, not this
recipe, which continues to run the pinned MinIO sidecar.
- **nginx 1.31.7** — fixes five CVEs (CVE-2026-15176, CVE-2026-93479, CVE-2026-95101,
CVE-2026-97629, CVE-2026-97205). Recommended.
- **redis 8.10.2** — security patch release.
- **collabora 26.04.x** — the 26.04 image executes `coolwsd` directly (no shell in the image):
the old `entrypoint`/`extra_params` wiring is replaced by `command` args and a `coolwsd --probe`
healthcheck; the admin-panel password moves from the retired Docker secret `collabora_p` to the
`COLLABORA_ADMIN_PASSWORD` env var (empty/unset ⇒ panel stays locked). Existing deployments are
unaffected by the orphaned `collabora_p` secret.
- **`minio_initialize` v1 → v2** — Swarm configs are immutable; the config content changed in an
earlier upstream change but the vendored version was left at `v1`, so upgrades of existing
deployments aborted with `only updates to Labels are allowed`. Bumping to `v2` makes Swarm create
a fresh config object.
### Advisory scan (step 2a)
Deterministic scan for `v0.22.0→v0.24.0` (+ redis/nginx/collabora) reports **5 CVEs fixed**, all
nginx (the five above). ⚠️ The scan's GitHub Security Advisories source returned HTTP 403
(rate-limited from the scan host) for every repo during this run, so that count is a **floor** — an
incomplete scan, not a clean one. redis 8.10.2's release notes carry security fixes without CVE ids;
collabora 26.04.4.2.1 (carried from the earlier run of this PR) resolves five GHSA advisories fixed
in the 25.04.11+/26.04.2+ lines.
### Release command (operator)
The recipe version label (`0.13.2+v0.22.0`) is intentionally **not** bumped here. After merge,
publish with:
```
abra recipe release lasuite-drive -y
```
**Not merged** — awaiting operator review.
Tested green on the cc-ci recipe CI server (full suite, cold, against this PR head). NOT merged — for operator review.
cc @trav @notplants
collabora 26.04 images exec coolwsd directly (--use-env-vars; no shell in the
image), so the sh/start-collabora-online.sh entrypoint and extra_params env are
gone: ssl options move to command args, the healthcheck switches to
coolwsd --probe, and the admin panel password moves from the collabora_p
docker secret to the COLLABORA_ADMIN_PASSWORD env var (empty = locked).
redis 8.10.2 is a security release (transaction ACL-revocation bypass,
unauthenticated cluster-bus join; cache sidecar, no cluster mode here).
autonomic-bot
requested review from trav 2026-09-21 21:28:34 +00:00
autonomic-bot
requested review from notplants 2026-09-21 21:28:34 +00:00
autonomic-bot
changed title from chore: upgrade lasuite-drive redis to 8.10.2, collabora to 26.04.4.1.1 to chore: upgrade lasuite-drive to v0.23.0, collabora to 26.04.4.2.1; bump minio_initialize config to v22026-09-28 21:07:18 +00:00
autonomic-bot
changed title from chore: upgrade lasuite-drive to v0.23.0, collabora to 26.04.4.2.1; bump minio_initialize config to v2 to chore: upgrade lasuite-drive to v0.24.0, redis to 8.10.2, nginx to 1.31.7; collabora 26.04.4.2.12026-10-16 03:00:39 +00:00
!testme run 60 — RED at the upgrade BASE (external blocker, not this upgrade, not a test).
The harness resolved the upgrade base as the last-green data-warm canonical 0.12.0+v0.21.0 (warm-lasuite-drive.ci.commoninternet.net, idle, ts 2026-09-13) and its
pre-deploy prepull tried to pull that version's MinIO image, minio/minio:RELEASE.2025-09-07T16-13-09Z — a Docker Hub repo removed upstream:
prepull: pulling minio/minio:RELEASE.2025-09-07T16-13-09Z …
!! deploy/readiness failed: `docker pull minio/minio:RELEASE.2025-09-07T16-13-09Z` failed (rc=1)
— Error response from daemon: pull access denied for minio/minio, repository does not exist
install: fail (upgrade/backup/restore/custom: skip — the run never reached them)
This PR's head pins bitnamilegacy/minio:2025.7.23-debian-12-r5 (the upstream repoint) and is not the failing image. The local image store no longer caches the removed minio/minio tag
(it was present when this PR last ran green on 2026-09-28), so the base deploy now fails
deterministically before the upgrade tier runs.
Evidence the upgrade itself is correct (step 2b, direct --chaos deploy of this head on cc-ci):
11/11 services healthy (drive v0.24.0, redis 8.10.2, nginx 1.31.7, collabora 26.04.4.2.1, minio_initialize config v2), backend migrations applied (migrate: done), collabora 26.04 WSD initialization complete, web 200, redis PONG — torn down afterwards (0 leaks).
This is the same external base blocker the sibling lasuite-docs PR #9 hit today (its warm
canonical also prepulls the removed docker.io/minio/minio). It is a canonical/base-image issue, not a recipe regression and not a stale test assertion — no test was modified (DEFAULT mode).
Operator options to unblock: refresh/rebuild the data-warm canonical to a post-repoint version
(≥ 0.13.0+v0.22.0, which pins bitnamilegacy/minio), or add a tests/lasuite-drive/previous/
overlay (VERSION marker 0.12.0+v0.21.0) overriding the base's MinIO image. Re-running /recipe-upgrade lasuite-drive --with-tests would let the upgrader add that overlay.
1 / 3 !testme runs used — not spending the remaining budget on a deterministic external failure. Nothing merged.
<!-- cc-ci:upgrade-note -->
**`!testme` run 60 — RED at the upgrade BASE (external blocker, not this upgrade, not a test).**
The harness resolved the upgrade **base** as the last-green data-warm canonical
`0.12.0+v0.21.0` (`warm-lasuite-drive.ci.commoninternet.net`, idle, ts 2026-09-13) and its
pre-deploy **prepull** tried to pull that version's MinIO image,
`minio/minio:RELEASE.2025-09-07T16-13-09Z` — a Docker Hub repo **removed upstream**:
```
prepull: pulling minio/minio:RELEASE.2025-09-07T16-13-09Z …
!! deploy/readiness failed: `docker pull minio/minio:RELEASE.2025-09-07T16-13-09Z` failed (rc=1)
— Error response from daemon: pull access denied for minio/minio, repository does not exist
install: fail (upgrade/backup/restore/custom: skip — the run never reached them)
```
This PR's head pins `bitnamilegacy/minio:2025.7.23-debian-12-r5` (the upstream repoint) and is
**not** the failing image. The local image store no longer caches the removed `minio/minio` tag
(it was present when this PR last ran green on 2026-09-28), so the base deploy now fails
deterministically before the upgrade tier runs.
**Evidence the upgrade itself is correct** (step 2b, direct `--chaos` deploy of this head on cc-ci):
11/11 services healthy (drive `v0.24.0`, redis `8.10.2`, nginx `1.31.7`, collabora `26.04.4.2.1`,
`minio_initialize` config `v2`), backend migrations applied (`migrate: done`), collabora 26.04
`WSD initialization complete`, web `200`, redis `PONG` — torn down afterwards (0 leaks).
This is the **same external base blocker the sibling lasuite-docs PR #9 hit today** (its warm
canonical also prepulls the removed `docker.io/minio/minio`). It is a canonical/base-image issue,
**not a recipe regression and not a stale test assertion** — no test was modified (DEFAULT mode).
**Operator options to unblock:** refresh/rebuild the data-warm canonical to a post-repoint version
(≥ `0.13.0+v0.22.0`, which pins `bitnamilegacy/minio`), or add a `tests/lasuite-drive/previous/`
overlay (VERSION marker `0.12.0+v0.21.0`) overriding the base's MinIO image. Re-running
`/recipe-upgrade lasuite-drive --with-tests` would let the upgrader add that overlay.
1 / 3 `!testme` runs used — not spending the remaining budget on a deterministic external failure.
**Nothing merged.**
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Upgrade lasuite-drive to drive v0.24.0
Extends the existing upgrade PR (branch
upgrade-ce5308c) with the drive quartetv0.22.0 → v0.24.0and nginx1.31.6 → 1.31.7, on top of the changes already carriedhere (redis 8.10.2, collabora 26.04.4.2.1,
minio_initializeconfig v2). Rebased onto thecurrent upstream
main(863691f): the earlier branch predated upstream'srenovate.json,MAINTENANCE.md, and maintainer-metadata changes, which are now preserved (not deleted).Image-tag diff vs upstream
mainlasuite/drive-frontendv0.22.0v0.24.0lasuite/drive-backendv0.22.0v0.24.0redis8.10.18.10.2nginx1.31.61.31.7collabora/code25.04.10.3.126.04.4.2.1(carried from this PR)minio_initializev1v2(carried; required for the Swarm config rename)Upstream release notes (one per changed service):
v0.22.0→v0.24.0: https://github.com/suitenumerique/drive/releases/tag/v0.24.0 (and https://github.com/suitenumerique/drive/releases/tag/v0.23.0)8.10.1→8.10.2: https://github.com/redis/redis/releases/tag/8.10.21.31.6→1.31.7: https://nginx.org/en/CHANGES25.04.10.3.1→26.04.4.2.1: https://www.collaboraonline.com/collabora-online-26-04-release-notes/Operator action required
automatically on backend startup (verified live: all
core.*,easy_thumbnails.*,malware_detection.*,rest_framework_api_key.*,sessions.*,sites.*migrations applied,migrate: done). Notable upstream changes are additive (rate-limiting on item creation,storage-quota admin, WOPI sandboxed downloads, item-list speedups). Upstream's own dev compose
switched its local object storage from MinIO to RustFS — that is upstream's compose, not this
recipe, which continues to run the pinned MinIO sidecar.
CVE-2026-97629, CVE-2026-97205). Recommended.
coolwsddirectly (no shell in the image):the old
entrypoint/extra_paramswiring is replaced bycommandargs and acoolwsd --probehealthcheck; the admin-panel password moves from the retired Docker secret
collabora_pto theCOLLABORA_ADMIN_PASSWORDenv var (empty/unset ⇒ panel stays locked). Existing deployments areunaffected by the orphaned
collabora_psecret.minio_initializev1 → v2 — Swarm configs are immutable; the config content changed in anearlier upstream change but the vendored version was left at
v1, so upgrades of existingdeployments aborted with
only updates to Labels are allowed. Bumping tov2makes Swarm createa fresh config object.
Advisory scan (step 2a)
Deterministic scan for
v0.22.0→v0.24.0(+ redis/nginx/collabora) reports 5 CVEs fixed, allnginx (the five above). ⚠️ The scan's GitHub Security Advisories source returned HTTP 403
(rate-limited from the scan host) for every repo during this run, so that count is a floor — an
incomplete scan, not a clean one. redis 8.10.2's release notes carry security fixes without CVE ids;
collabora 26.04.4.2.1 (carried from the earlier run of this PR) resolves five GHSA advisories fixed
in the 25.04.11+/26.04.2+ lines.
Release command (operator)
The recipe version label (
0.13.2+v0.22.0) is intentionally not bumped here. After merge,publish with:
Not merged — awaiting operator review.
Tested green on the cc-ci recipe CI server (full suite, cold, against this PR head). NOT merged — for operator review.
cc @trav @notplants
!testme
🌻 cc-ci —
lasuite-drive@ce5308c7❌ failurefull logs · dashboard
chore: upgrade lasuite-drive redis to 8.10.2, collabora to 26.04.4.1.1to chore: upgrade lasuite-drive to v0.23.0, collabora to 26.04.4.2.1; bump minio_initialize config to v2!testme
🌻 cc-ci —
lasuite-drive@1ed910df✅ passedfull logs · dashboard
chore: upgrade lasuite-drive to v0.23.0, collabora to 26.04.4.2.1; bump minio_initialize config to v2to chore: upgrade lasuite-drive to v0.24.0, redis to 8.10.2, nginx to 1.31.7; collabora 26.04.4.2.1!testme
🌻 cc-ci —
lasuite-drive@5a38a20c❌ failurefull logs · dashboard
!testmerun 60 — RED at the upgrade BASE (external blocker, not this upgrade, not a test).The harness resolved the upgrade base as the last-green data-warm canonical
0.12.0+v0.21.0(warm-lasuite-drive.ci.commoninternet.net, idle, ts 2026-09-13) and itspre-deploy prepull tried to pull that version's MinIO image,
minio/minio:RELEASE.2025-09-07T16-13-09Z— a Docker Hub repo removed upstream:This PR's head pins
bitnamilegacy/minio:2025.7.23-debian-12-r5(the upstream repoint) and isnot the failing image. The local image store no longer caches the removed
minio/miniotag(it was present when this PR last ran green on 2026-09-28), so the base deploy now fails
deterministically before the upgrade tier runs.
Evidence the upgrade itself is correct (step 2b, direct
--chaosdeploy of this head on cc-ci):11/11 services healthy (drive
v0.24.0, redis8.10.2, nginx1.31.7, collabora26.04.4.2.1,minio_initializeconfigv2), backend migrations applied (migrate: done), collabora 26.04WSD initialization complete, web200, redisPONG— torn down afterwards (0 leaks).This is the same external base blocker the sibling lasuite-docs PR #9 hit today (its warm
canonical also prepulls the removed
docker.io/minio/minio). It is a canonical/base-image issue,not a recipe regression and not a stale test assertion — no test was modified (DEFAULT mode).
Operator options to unblock: refresh/rebuild the data-warm canonical to a post-repoint version
(≥
0.13.0+v0.22.0, which pinsbitnamilegacy/minio), or add atests/lasuite-drive/previous/overlay (VERSION marker
0.12.0+v0.21.0) overriding the base's MinIO image. Re-running/recipe-upgrade lasuite-drive --with-testswould let the upgrader add that overlay.1 / 3
!testmeruns used — not spending the remaining budget on a deterministic external failure.Nothing merged.
View command line instructions
Checkout
From your project repository, check out a new branch and test the changes.