On a brand new RHEL 10.1 installation, configured as a server with headless management tools, all Podman containers randomly exit, and do not start again. The Podman API Service also stops with shutdown.Stop(), but it was my understanding that Podman didn’t require a daemon. However, restarting the Podman service makes all services restart, here is an excerpt from journalctl -u podman for the past two runs:
Nov 26 12:01:45 earth systemd[1]: Started podman.service - Podman API Service.
Nov 26 12:01:45 earth podman[43676]: time="2025-11-26T12:01:45-07:00" level=info msg="/usr/bin/podman filtering at log level info"
Nov 26 12:01:45 earth podman[43676]: time="2025-11-26T12:01:45-07:00" level=info msg="Setting parallel job count to 13"
Nov 26 12:01:45 earth podman[43676]: time="2025-11-26T12:01:45-07:00" level=info msg="Using sqlite as database backend"
Nov 26 12:01:45 earth podman[43676]: time="2025-11-26T12:01:45-07:00" level=info msg="Not using native diff for overlay, this may cause degraded performance for building images: kernel has CONFIG_OVERLAY_FS_REDIRECT_DIR enabled"
Nov 26 12:01:45 earth podman[43676]: time="2025-11-26T12:01:45-07:00" level=info msg="Using systemd socket activation to determine API endpoint"
Nov 26 12:01:45 earth podman[43676]: time="2025-11-26T12:01:45-07:00" level=info msg="API service listening on "/run/podman/podman.sock". URI: "unix:///run/podman/podman.sock""
Nov 26 12:01:50 earth podman[43676]: time="2025-11-26T12:01:50-07:00" level=info msg="Received shutdown.Stop(), terminating!" PID=43676
Nov 26 12:01:50 earth systemd[1]: podman.service: Deactivated successfully.
Nov 26 12:01:57 earth systemd[1]: Starting podman.service - Podman API Service...
Nov 26 12:01:57 earth systemd[1]: Started podman.service - Podman API Service.
Nov 26 12:01:57 earth podman[43809]: time="2025-11-26T12:01:57-07:00" level=info msg="/usr/bin/podman filtering at log level info"
Nov 26 12:01:57 earth podman[43809]: time="2025-11-26T12:01:57-07:00" level=info msg="Setting parallel job count to 13"
Nov 26 12:01:57 earth podman[43809]: time="2025-11-26T12:01:57-07:00" level=info msg="Using sqlite as database backend"
Nov 26 12:01:57 earth podman[43809]: time="2025-11-26T12:01:57-07:00" level=info msg="Not using native diff for overlay, this may cause degraded performance for building images: kernel has CONFIG_OVERLAY_FS_REDIRECT_DIR enabled"
Nov 26 12:01:57 earth podman[43809]: time="2025-11-26T12:01:57-07:00" level=info msg="Using systemd socket activation to determine API endpoint"
Nov 26 12:01:57 earth podman[43809]: time="2025-11-26T12:01:57-07:00" level=info msg="API service listening on "/run/podman/podman.sock". URI: "unix:///run/podman/podman.sock""
Nov 26 12:01:57 earth podman[43809]: @ - - [26/Nov/2025:12:01:57 -0700] "GET /v3.4.0/libpod/info? HTTP/1.1" 200 3213 "" ""
Nov 26 12:01:57 earth podman[43809]: @ - - [26/Nov/2025:12:01:57 -0700] "GET /v3.4.0/libpod/images/json? HTTP/1.1" 200 3 "" ""
Nov 26 12:01:57 earth podman[43809]: @ - - [26/Nov/2025:12:01:57 -0700] "GET /v3.4.0/libpod/containers/json?all=true HTTP/1.1" 200 3 "" ""
Nov 26 12:01:57 earth podman[43809]: @ - - [26/Nov/2025:12:01:57 -0700] "GET /v3.4.0/libpod/pods/json? HTTP/1.1" 200 3 "" ""
Nov 26 12:02:22 earth podman[43809]: @ - - [26/Nov/2025:12:01:57 -0700] "GET /v3.4.0/libpod/events HTTP/1.0" 200 0 "" ""
Nov 26 12:02:27 earth podman[43809]: @ - - [26/Nov/2025:12:01:57 -0700] "GET /v3.4.0/libpod/containers/stats HTTP/1.0" 200 156 "" ""
Nov 26 12:02:32 earth podman[43809]: time="2025-11-26T12:02:32-07:00" level=info msg="Received shutdown.Stop(), terminating!" PID=43809
Nov 26 12:02:32 earth systemd[1]: podman.service: Deactivated successfully.
All containers where setup using docker-compose (through podman compose) and have restart: unless-stopped. There server is up-to-date and there are is nothing from SELinux (other than dnf not being able to read the tailscale.repo file.)
The containers don’t mention being shutdown in their logs, they just exit.
