during boot ,some container failed ,lots of config not applied .need to load and commit . is it able to bypass container failure ?
Are you sure it’s the container failing that’s cause config to not apply?
What version of VyOS are you using?
one container has some volume mount issue , then all containers were blocked .even some service cannot be applied during boot .I have to load and delete the container and commit again .
vyos version is rolling ,but this issue seems exited for a longtime . no one else met this kind of issue ?
Rolling from which date, yesterday or 2 years old rolling?
Can you paste your container-config using “show config commands”?
I encountered the same issue on VyOS Stream 2026.03 when I had 2 containers and one failed. I was able to get a more useful error message by running sudo journalctl -xe and scrolling up to find the container’s logs. In my case the error message was along the lines of Error: /big/long/path/ending/in/merged. Directory not empty and I was able to temporarily solve it by deleting /big/long/path/ending/in/merged but it tended to recur after a few power-cycles.
The containers I was running were mietzen/porkbun-ddns:latest and mvance/unbound:latest. The Porkbun-DDNS container was the only one that created errors, but when the errors did happen, the Unbound container wouldn’t get loaded (and my home network’s DNS resolver would be down as a result).
I cannot comment on whether the latest rolling release solves this issue, as the latest rolling release adds porkbun support to the built-in dynamic dns updating, negating the need for the problematic porkbun container, which I deleted.