-
fix(docker): degrade to summary-only details when inspects time out; best-effort snapshots
Release / frontend (push) Canceled after 0sRelease / linux (amd64) (push) Canceled after 0sRelease / linux (arm64) (push) Canceled after 0sRelease / windows (push) Canceled after 0sRelease / macos (push) Canceled after 0sRelease / release (push) Canceled after 0sBuild and Push Docker Image / build-and-push (push) Canceled after 0sreleased this
2026-06-12 12:48:45 +03:00 On hosts with a wedged-slow docker daemon (dokploy VM) every
ContainerInspect exceeds its 5s slice even after the sized-list fix.
Two consequences fixed:-
the collector dropped each uninspectable container entirely, so the
Applications view stayed empty exactly where the operator needed it.
Inspect only adds marginal fields (health, finished-at, cpu-limit,
precise created, config image) on top of the list summary — failed
inspects now fall back to summary-built containers, and the per-
container log flood collapses into one WARN per cycle; -
CollectAllCurrent returned an error when ANY module failed, so the
SSE envelope AND the cluster metric batch were suppressed wholesale —
a slow docker daemon silenced the node's cpu/mem/disk/net on every
peer and the hub showed '--'. The snapshot is now best-effort:
failed modules are omitted (not nulled — SSE consumers merge
per-key), everything else ships.
Downloads
-