https://bz.apache.org/bugzilla/show_bug.cgi?id=65402
--- Comment #9 from Stefan Eissing <ste...@eissing.org> --- > Yes. That's me clicking around in the app. After every route change the 2 > Eventsources are destroyed and connected again. (I know that's not how it > should be - we are working on it to persist the event-sources during > route-changes in the app) I see. Ok that explains this. > That's true. When connecting to the SSE endpoint there is some initial > payload - but then we are sending updates only every 30? seconds or when some > change is happening. The closed connections is me navigating through the page > (as described above) Ok, if the SSE source sends roughly every 30 seconds, this explains what I see: - the browser closes - mod_http2 acknowledges, and marks the worker thread as being aborted - the worker thread is in a blocking read and gets some data after 30 seconds, detects the frontend connection is gone and terminates itself. > Can you clarify that for me - I don't understand the part with "the browser > detects frequent data ..." That was my question about the early closing of the SSE requests. But you explained that you navigated the webapp and caused this yourself. What I do not see is any "blocking" behaviour in the server. If you see a request in the browser dev tools that is not answered, you should see the URL of that request and then we can check if we see it in the server logs and what is happening to it. For the SSE source requests, the browser will most likely show a "waiting" status since the complete response is never received - as intended for the SSE, right? -- You are receiving this mail because: You are the assignee for the bug. --------------------------------------------------------------------- To unsubscribe, e-mail: bugs-unsubscr...@httpd.apache.org For additional commands, e-mail: bugs-h...@httpd.apache.org
