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
Description
I recently upgraded from 3006.9 to 3006.10, and a new memory leak appears to have been introduced.
I now have to restart salt-master basically every day to avoid OOM issues, and will probably have to revert to .9
Setup
Master is on KVM. I have only twelve minions, on a mix of machine, but all the same version as the master.
They run state.apply every fifteen minutes, and a couple run some other scheduled script that doesn't interact with the master.
Both gitfs and git_pillar are in use, as well as x509 (v1) states and mine requests.
Steps to Reproduce the behavior
Salt process shortly after a restart:
Description
I recently upgraded from 3006.9 to 3006.10, and a new memory leak appears to have been introduced.
I now have to restart
salt-master
basically every day to avoid OOM issues, and will probably have to revert to .9Setup
Master is on KVM. I have only twelve minions, on a mix of machine, but all the same version as the master.
They run
state.apply
every fifteen minutes, and a couple run some other scheduled script that doesn't interact with the master.Both gitfs and git_pillar are in use, as well as x509 (v1) states and
mine
requests.Steps to Reproduce the behavior
Salt process shortly after a restart:
Salt processes thirty minutes later:
24 hours later:
Expected behavior
Memory usage should not keep growing.
Versions Report
salt --versions-report
Additional information
Version report from before the upgrade is here: #67207
The text was updated successfully, but these errors were encountered: