Syx
Syx
same problem https://github.com/kubernetes/node-problem-detector/issues/726
@Random-Liu @wangzhen127
@miekg @mtharp
ound two related commits that introduce the use of dlopen: https://github.com/miekg/pkcs11/pull/102 https://github.com/miekg/pkcs11/pull/28
I added some log information. Under normal circumstances, once crio.xxx.scope receives an exit message(systemctl kill crio-xxx.scope), conmon should be able to capture it, just like this: ```bash Feb 24 09:37:54...
However, under abnormal circumstances, after crio-xxx.scope exits, conmon does not receive any signals. Journalctl log ```bash Feb 24 10:05:14 node71 crio[4983]: time="2025-02-24 10:05:14.387786482+08:00" level=info msg="running conmon: /usr/bin/conmon" args="[-b /run/containers/storage/overlay-containers/c50666e24d9091a348a831015800cd8c603e3fb9ca23958b249f74e17487f21d/userdata -c...
> Thanks for the detailed report and logs, [@Syx5290](https://github.com/Syx5290)! You've identified a real issue with conmon's container exit detection in certain systemd configurations. Thanks for confirming the issue and for...
Can you provide any clues to help me analyze this issue? Thank you. @haircommander
> `cgroup: subsystem does not exist` and `error jailing process inside rootfs: pivot_root` seem like errors coming from runc cc [@kolyshkin](https://github.com/kolyshkin) Similar like this : https://github.com/docker/for-linux/issues/1127
> is this reproducible? the root cause of the issue has not been identified and it cannot be reproduced, although it occurs quite frequently. I am doing my best to...