Page MenuHomeVyOS Platform

Podman health checks stay in starting
Closed, ResolvedPublicBUG

Description

On VyOS rolling 2026.08.22-0013 with Podman 5.8.4, container health checks never run automatically and stay in starting.

Current config example:

health-check {
    command "openresty -t -q"
    interval 30
    retry 3
    timeout 5
}

Podman also receives the expected options:

--health-cmd=openresty -t -q
--health-interval=30s
--health-timeout=5s
--health-retries=3

Manual checks work fine:

podman healthcheck run openresty

After that, the container becomes healthy.

But systemctl list-timers --all shows no Podman health-check timers.

It looks like Podman may have been built without the systemd build tag.

Possible fix: add libsystemd-dev to the Podman build dependencies so automatic health-check scheduling is enabled.

Details

Version
2026.08.22-0013
Is it a breaking change?
Unspecified (possibly destroys the router)
Issue type
Bug (incorrect behavior)

Event Timeline

Podman upstream has an open PR related to exactly this case: https://github.com/podman-container-tools/podman/pull/27033 ("healthcheck for non-systemd podman"). The current non-systemd implementation doesn't schedule health checks automatically, and that PR adds a scheduler for it.

For reference, journald logging still works on VyOS because conmon itself is linked with libsystemd:

libsystemd.so.0 => /lib/x86_64-linux-gnu/libsystemd.so.0

So working log-driver journald doesn't necessarily mean Podman itself was built with the systemd build tag.

This makes me suspect that the VyOS Podman package is being built without systemd support. Adding libsystemd-dev to the Podman build dependencies should allow Podman's hack/systemd_tag.sh to enable the systemd build tag.

Viacheslav triaged this task as Normal priority.Wed, Aug 26, 4:20 PM
Viacheslav changed the task status from Open to In progress.Fri, Aug 28, 8:25 AM
Viacheslav assigned this task to opswill.
Viacheslav moved this task from Need Triage to Completed on the VyOS Rolling board.