https://bugs.winehq.org/show_bug.cgi?id=54052
Bug ID: 54052 Summary: winhttp:notification times out randomly in Wine Product: Wine Version: unspecified Hardware: x86-64 OS: Linux Status: NEW Severity: normal Priority: P2 Component: winhttp Assignee: wine-bugs@winehq.org Reporter: fgouget@codeweavers.com Distribution: ---
winhttp:notification times out randomly in Wine.
First, on 2022-11-24 Paul Gofman got a series of patches in that probably fixed a bunch of these random timeouts (the situation on fg-deb64 and fgtb-debian11 seems better for one).
However the following merge requests got timeouts after the patches went in:
* MR!1553 on 2022-11-26 ... notification.c:117: Test failed: 892: expected status 0x20000 got 0x200000 winhttp:notification:0b98 done (258) in 120s 1148B
* MR!1557 on 2022-11-27 ... notification.c:117: Test failed: 988: expected status 0x80000 got 0x200000 winhttp:notification:06d0 done (258) in 120s 1148B
* MR!1560 on 2022-11-27 ... notification.c:1184: Test marked todo: got 3 winhttp:notification:052c done (258) in 120s 1118B
Where: 0x20000 == WINHTTP_CALLBACK_STATUS_HEADERS_AVAILABLE 0x80000 == WINHTTP_CALLBACK_STATUS_READ_COMPLETE 0x200000 == WINHTTP_CALLBACK_STATUS_REQUEST_ERROR
So there may still be some issue left.
https://bugs.winehq.org/show_bug.cgi?id=54052
François Gouget fgouget@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |source, testcase CC| |pgofman@codeweavers.com
https://bugs.winehq.org/show_bug.cgi?id=54052
--- Comment #1 from Paul Gofman pgofman@codeweavers.com --- It gives request errors on receiving server response. I think that it is likely some network / server connection issues which give some flakiness here. At least for the first failing place we have a check in the prior WinHttpSendRequest to handle server unavailability gracefully. Error in receiving response is much less likely after that but probably can still happen.
I have MR 1582 (https://gitlab.winehq.org/wine/wine/-/merge_requests/1582) which, as a side effect, can improve this at least for the first failing case: after that MR receiving actual reply should happen in WinHttpSendRequest() so in case of errors there we should fall into the same server unavailability check.
So I think ideally instead of going for checking server intermittent failures in each of those requests in the test I'd wait for that commit to be in first and then look what's left.
https://bugs.winehq.org/show_bug.cgi?id=54052
--- Comment #2 from Hans Leidekker hans@meelstraat.net --- This looks better now. There are no Wine failures since the patches were merged about a month ago.
https://bugs.winehq.org/show_bug.cgi?id=54052
François Gouget fgouget@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution|--- |FIXED
--- Comment #3 from François Gouget fgouget@codeweavers.com --- Yes I think this is fixed. Marking it as such.
https://bugs.winehq.org/show_bug.cgi?id=54052
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #4 from Alexandre Julliard julliard@winehq.org --- Closing bugs fixed in 8.4.