After the last update, Dokuwiki no longer starts. Reconfiguration leads to:
Job for dokuwiki.service failed because the control process exited with error code.
See "systemctl --user status dokuwiki.service" and "journalctl --user -xeu dokuwiki.service" for details.
I wanted to check the satus and logs, so I tried to switch to the container.
But then I realized that the container no longer exists at all.
root@daho-ns8:~# runagent -m dokuwiki1 podman ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
root@daho-ns8:~#
NS8-Logs
2026-07-27T13:14:32+02:00 [1:dokuwiki1:systemd] Started libpod-98c4cd64f57e0b291607fa8fe6d4cfa868a7d71383de131a93e9ee3176e87956.scope - libcrun container.
2026-07-27T13:14:32+02:00 [1:dokuwiki1:podman] 2026-07-27 13:14:32.197505118 +0200 CEST m=+6.766147513 container init 98c4cd64f57e0b291607fa8fe6d4cfa868a7d71383de131a93e9ee3176e87956 (image=docker.io/dokuwiki/dokuwiki:2025-05-14b, name=dokuwiki, PODMAN_SYSTEMD_UNIT=dokuwiki.service)
2026-07-27T13:14:32+02:00 [1:dokuwiki1:podman] 2026-07-27 13:14:32.205985696 +0200 CEST m=+6.774628051 container start 98c4cd64f57e0b291607fa8fe6d4cfa868a7d71383de131a93e9ee3176e87956 (image=docker.io/dokuwiki/dokuwiki:2025-05-14b, name=dokuwiki, PODMAN_SYSTEMD_UNIT=dokuwiki.service)
2026-07-27T13:14:32+02:00 [1:dokuwiki1:dokuwiki] + set -e
2026-07-27T13:14:32+02:00 [1:dokuwiki1:dokuwiki] ++ id -u
2026-07-27T13:14:32+02:00 [1:dokuwiki1:podman] 98c4cd64f57e0b291607fa8fe6d4cfa868a7d71383de131a93e9ee3176e87956
2026-07-27T13:14:32+02:00 [1:dokuwiki1:dokuwiki] ++ id -g
2026-07-27T13:14:32+02:00 [1:dokuwiki1:dokuwiki] + echo 'Started under UID 0 GID 0.'
2026-07-27T13:14:32+02:00 [1:dokuwiki1:dokuwiki] Started under UID 0 GID 0.
2026-07-27T13:14:32+02:00 [1:dokuwiki1:dokuwiki] + '[' 0 -eq 0 ']'
2026-07-27T13:14:32+02:00 [1:dokuwiki1:dokuwiki] + echo 'Running as root, dropping privileges to 33:33'
2026-07-27T13:14:32+02:00 [1:dokuwiki1:dokuwiki] Running as root, dropping privileges to 33:33
2026-07-27T13:14:32+02:00 [1:dokuwiki1:dokuwiki] + chown -R www-data:www-data /storage
2026-07-27T13:14:32+02:00 [1:dokuwiki1:dokuwiki] chown: changing ownership of '/storage/data/pages/ns8': Operation not permitted
2026-07-27T13:14:32+02:00 [1:dokuwiki1:dokuwiki] chown: changing ownership of '/storage/data/pages/pages/start/nethserver8.txt': Operation not permitted
2026-07-27T13:14:32+02:00 [1:dokuwiki1:dokuwiki] chown: changing ownership of '/storage/data/pages/pages/start': Operation not permitted
2026-07-27T13:14:32+02:00 [1:dokuwiki1:dokuwiki] chown: changing ownership of '/storage/data/pages/pages': Operation not permitted
2026-07-27T13:14:32+02:00 [1:dokuwiki1:conmon] conmon 98c4cd64f57e0b291607 <nwarn>: Failed to open cgroups file: /sys/fs/cgroup/user.slice/user-1005.slice/user@1005.service/user.slice/libpod-98c4cd64f57e0b291607fa8fe6d4cfa868a7d71383de131a93e9ee3176e87956.scope/container/memory.events
2026-07-27T13:14:32+02:00 [1:dokuwiki1:podman] 2026-07-27 13:14:32.501762962 +0200 CEST m=+0.100698242 container died 98c4cd64f57e0b291607fa8fe6d4cfa868a7d71383de131a93e9ee3176e87956 (image=docker.io/dokuwiki/dokuwiki:2025-05-14b, name=dokuwiki, PODMAN_SYSTEMD_UNIT=dokuwiki.service)
2026-07-27T13:14:32+02:00 [1:dokuwiki1:podman] 2026-07-27 13:14:32.623241258 +0200 CEST m=+0.222176516 container cleanup 98c4cd64f57e0b291607fa8fe6d4cfa868a7d71383de131a93e9ee3176e87956 (image=docker.io/dokuwiki/dokuwiki:2025-05-14b, name=dokuwiki, PODMAN_SYSTEMD_UNIT=dokuwiki.service)
2026-07-27T13:14:32+02:00 [1:dokuwiki1:systemd] dokuwiki.service: Main process exited, code=exited, status=1/FAILURE
2026-07-27T13:14:32+02:00 [1:dokuwiki1:systemd] dokuwiki.service: Control process exited, code=exited, status=1/FAILURE
2026-07-27T13:14:33+02:00 [1:dokuwiki1:podman] time="2026-07-27T13:14:33+02:00" level=error msg="User-selected graph driver \"overlay\" overwritten by graph driver \"vfs\" from database - delete libpod local files (\"/home/dokuwiki1/.local/share/containers/storage\") to resolve. May prevent use of images created by other tools"
2026-07-27T13:14:34+02:00 [1:dokuwiki1:podman] 2026-07-27 13:14:34.131530325 +0200 CEST m=+1.099366197 container remove 98c4cd64f57e0b291607fa8fe6d4cfa868a7d71383de131a93e9ee3176e87956 (image=docker.io/dokuwiki/dokuwiki:2025-05-14b, name=dokuwiki, PODMAN_SYSTEMD_UNIT=dokuwiki.service)
2026-07-27T13:14:34+02:00 [1:dokuwiki1:podman] 98c4cd64f57e0b291607fa8fe6d4cfa868a7d71383de131a93e9ee3176e87956
2026-07-27T13:14:34+02:00 [1:dokuwiki1:systemd] dokuwiki.service: Failed with result 'exit-code'.
2026-07-27T13:14:34+02:00 [1:dokuwiki1:systemd] Failed to start dokuwiki.service - Dokuwiki instance.
2026-07-27T13:14:34+02:00 [1:dokuwiki1:systemd] dokuwiki.service: Consumed 9.730s CPU time, 59.8M memory peak.
2026-07-27T13:14:34+02:00 [1:dokuwiki1:systemd] dokuwiki.service: Scheduled restart job, restart counter is at 91.
2026-07-27T13:14:34+02:00 [1:dokuwiki1:systemd] Starting dokuwiki.service - Dokuwiki instance...
2026-07-27T13:14:36+02:00 [1:dokuwiki1:runagent] Configuration written to dokuwiki-config/local.protected.php
2026-07-27T13:14:36+02:00 [1:dokuwiki1:runagent] Configuration written to dokuwiki-config/plugins.local.php
2026-07-27T13:14:36+02:00 [1:dokuwiki1:podman] time="2026-07-27T13:14:36+02:00" level=error msg="User-selected graph driver \"overlay\" overwritten by graph driver \"vfs\" from database - delete libpod local files (\"/home/dokuwiki1/.local/share/containers/storage\") to resolve. May prevent use of images created by other tools"
2026-07-27T13:14:36+02:00 [1:dokuwiki1:podman] 2026-07-27 13:14:36.244394436 +0200 CEST m=+0.051426993 image pull 1d386be80cbcbcd9b38d3a7c188f6e5b32388316e69a2e05cc75405b856fa392 docker.io/dokuwiki/dokuwiki:2025-05-14b
grep dokuwiki1 /etc/subuid
ls -lnR /home/dokuwiki1/.local/share/containers/storage/volumes/dokuwiki-data/_data/data/pages/
The ns8/ and pages/ directories in your volume are rejecting the chown that DokuWiki’s entrypoint performs at startup, which is what makes the service fail. This happens when files are copied into the volume from the host (as root, or via rsync -a from a backup) instead of through the module’s namespace.
Did you import wiki content manually at some point? And was data/pages/pages/ intentional — it would map to a pages:start:nethserver8 page, which looks like a nesting mistake.
Did you import wiki content manually at some point? And was data/pages/pages/ intentional — it would map to a pages:start:nethserver8 page, which looks like a nesting mistake.
Yes, during Migration from NS7 to NS8 I imported the Wiki-Content. …some years ago.
Those are outside the module’s UID mapping, so inside the container they show up as nobody. DokuWiki’s entrypoint runs chown -R www-data:www-data /storage at every container start, and it runs with set -e — so the first Operation not permitted aborts the whole startup. That’s why the service won’t come up. It’s not caused by the upgrade itself; the upgrade just triggered the first restart since those files appeared on 15 July.