You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When I'm doing node-e2e test for Pouch CRI, there are lot of log like:
ERRO[2018-06-15 06:57:18.160883152] Error deleting network: failed to Statfs "/proc/10435/ns/net": no such file or directory
ERRO[2018-06-15 06:57:18.160936276] StopPodSandbox for "0ad2446f4a41ad160ac10b59a3efeb81a284a955decbee74cb37750d5a303073" failed, error: failed to destroy network for sandbox "0ad2446f4a41ad160ac10b59a3efeb81a284a955decbee74cb37750d5a303073": failed to Statfs "/proc/10435/ns/net": no such file or directory
That means the pod can't be stopped because of the pod network can't be teared down.
However through pouch ps -a, the corresponding container has been stopped, so we can't find the corresponding network namespace obviously and we won't stop the sandbox successfully forever.
So I need to configure now whether kubelet will stop sandbox multiple times, if it is true, we need to change the logic of StopSandbox in CRI manager.
The text was updated successfully, but these errors were encountered:
Ⅰ. Issue Description
When I'm doing node-e2e test for Pouch CRI, there are lot of log like:
That means the pod can't be stopped because of the pod network can't be teared down.
However through
pouch ps -a
, the corresponding container has been stopped, so we can't find the corresponding network namespace obviously and we won't stop the sandbox successfully forever.So I need to configure now whether kubelet will stop sandbox multiple times, if it is true, we need to change the logic of
StopSandbox
in CRI manager.The text was updated successfully, but these errors were encountered: