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 we bumped the library from v1.8.7 to v1.8.10 we started seeing huge memory usage spikes when the process is shutting down. It turned out to be because the library starts lots of goroutines in some (?) case.
Blue graph is RAM usage in percentage of max allowed. Note how spikes start at some point - that's when the version with the library update was rolled out.
Screenshot from Google Profiler with the width of the flame graph showing the proportion of goroutines with a certain stacktrace. Note that at the moment of capture there was 153,983 goroutines, most of them from the library.
This is a regression. Code worked fine until the version bump and works fine again since we downgraded the library.
When we bumped the library from
v1.8.7
tov1.8.10
we started seeing huge memory usage spikes when the process is shutting down. It turned out to be because the library starts lots of goroutines in some (?) case.Blue graph is RAM usage in percentage of max allowed. Note how spikes start at some point - that's when the version with the library update was rolled out.

Screenshot from Google Profiler with the width of the flame graph showing the proportion of goroutines with a certain stacktrace. Note that at the moment of capture there was 153,983 goroutines, most of them from the library.
This is a regression. Code worked fine until the version bump and works fine again since we downgraded the library.
Original issue: https://gitlab.com/gitlab-org/cluster-integration/gitlab-agent/-/issues/492
The text was updated successfully, but these errors were encountered: