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
We see errors for requests that were scheduled in race conditions with GOAWAY streams.
Expected behavior
If a request was scheduled on an http2 stream with a streamID higher than the lastStreamID of the GOAWAY frame, then this request is save to rerun. We should do this.
The text was updated successfully, but these errors were encountered:
Just want to give an update about the original issue (#488) after ~ a year and a half.
The update is ... Since yesterday I've enabled back http2 requests and this time the server doesn't start to fail requests after a few hours (which used to be the case).
However, one request fails every once in a while with a stream-closed error:
The timings of those errors is not consistent but i can see a pattern. The errors happen every 20 or multitude of 20 minutes. So sometimes after 20 minutes, sometimes after an hour, up to maybe, let's say, 2 hours.
The stream ids are also always very close to 20000 (20001/20003).
Follow up from: #488
Observed behavior
We see errors for requests that were scheduled in race conditions with GOAWAY streams.
Expected behavior
If a request was scheduled on an http2 stream with a streamID higher than the lastStreamID of the GOAWAY frame, then this request is save to rerun. We should do this.
The text was updated successfully, but these errors were encountered: