Replies: 1 comment 1 reply
-
It pains me to say this, but it is a known limitation. |
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hello, community.
I'm creating 600 pods with PVs in parallel (this is the requirement of one of our clients). I'm trying to understand when the allocated size of the node will be updated.
As far as I can tell, there is a VG service that runs once every ten minutes and notifies others about updated status
topolvm/internal/lvmd/embed.go
Lines 35 to 46 in ecfadb5
Can the scheduler accidentally mispredict the node because I already allocated all the space, but the notifier still hasn't finished the cycle and there are still outdated values for allocated capacity on a node?
Beta Was this translation helpful? Give feedback.
All reactions