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
It seems that the application's qps is often higher than usual, this might result in generating larger responses for /metrics, when vmagent has high memory usage.
I will share the profile here if it happens again.
Not sure I should create a separate thread, but we're seeing that 6 of ~30 vmagents have seen the memory usage double when upgrading from v1.93.14 to v1.101.0, running the same chart and operator versions. No difference in target count. pprof.zip
Is your question request related to a specific component?
vmagent
Describe the question in detail
I find VMAgent got OOMKilled after upgrading from v1.79 -> v1.97
and its memory usage fluctuate significantly, what might be the cause?
Troubleshooting docs
The text was updated successfully, but these errors were encountered: