Cannot start sandbox pod about cri-o HOT 7 CLOSED

@14rcole I think that adding syslog support to conmon will help diagnose this issue. conmon is probably exiting without sending pid back to ocid.

from cri-o.

@14rcole Meanwhile to cleanup: runc delete -f all running containers, rm -rf /var/lib/ocid and then restart ocid server.

from cri-o.

@mrunalp I tried cleaning up , I'm getting the same error. I think it does have to do with conmon. On the daemon side I'm getting the messages

conmon: Failed to create container INFO[0007] Signal received: child exited INFO[0007] Reaped process with pid 30839

That "failed to start container" message comes from conmon.c line 164. It looks like system(cmd) is returning the error code 768

from cri-o.

@14rcole What version of runc are you using? It should be > 1.0.0.rc1.

from cri-o.

I was using the one in the dnf repositories, which was 0.1.1. After building the latest release on github, it works fine. Thank you!

from cri-o.

Please close this issue.

from cri-o.

@14rcole Thanks for the update!

from cri-o.