Werk #19463: Podman: containers exited with code 0 are removed from individual monitoring
| Component | Checks & agents | ||||
| Title | Podman: containers exited with code 0 are removed from individual monitoring | ||||
| Date | May 12, 2026 | ||||
| Level | Prominent Change | ||||
| Class | New Feature | ||||
| Compatibility | Incompatible - Manual interaction might be required | ||||
| Checkmk versions & editions |
|
The Podman agent plug-in (mk_podman) now changes how exited containers are
handled in individual container monitoring (piggyback hosts).
Containers that have exited with exit code 0 (clean exit) are no longer monitored individually. Their piggyback host is removed, which causes all associated services to disappear from monitoring at the next check cycle.
A new plug-in configuration option, Keep containers exited with non-zero exit code in monitoring, controls whether containers that exited with a non-zero exit code remain individually monitored. This option is enabled by default, preserving the previous behaviour for failed containers.
Note: all exited containers (exit code 0 or non-zero) continue to be counted in the Podman containers service on the Podman host, regardless of these settings.
To configure this option, go to Setup > Agents > Windows, Linux, Solaris, AIX and edit the rule Podman hosts and containers (Linux).