Closed cyclinder closed 1 month ago
why to delay restarting agent ? the problem is the confimap is not updated. Ss there any mechanism in the agent writing the agent ?
I mean we should be checking the configMap if updated first, and then restart the agent.
Spiderpool Version
all
Bug Type
Other
Main CNI
all
What happened?
We hit the issue about pods can't be started due to no default cni config found, we found the clusterNetwork of configMap is empty, but all logs show works.
What did you expect to happen?
init-pod: restart agent should wait the multus configMap updated.
How to reproduce it (as minimally and precisely as possible)
Additional Context
No response