Summary: | REGRESSION(r206731): [SOUP] Network process crash in gotHeadersCallback | ||||||
---|---|---|---|---|---|---|---|
Product: | WebKit | Reporter: | Michael Catanzaro <mcatanzaro> | ||||
Component: | WebKitGTK | Assignee: | Nobody <webkit-unassigned> | ||||
Status: | RESOLVED FIXED | ||||||
Severity: | Normal | CC: | berto, bugs-noreply, cgarcia, commit-queue, danw, gustavo, mcatanzaro, mrobinson | ||||
Priority: | P2 | ||||||
Version: | Other | ||||||
Hardware: | PC | ||||||
OS: | Linux | ||||||
Attachments: |
|
Description
Michael Catanzaro
2016-10-08 16:01:24 PDT
I believe it introduced crashes in the following tests (including all the HLS tests): http/tests/download/area-download.html http/tests/media/hls/hls-audio-tracks-has-audio.html http/tests/media/hls/hls-audio-tracks-locale-selection.html http/tests/media/hls/hls-audio-tracks.html http/tests/media/hls/hls-progress.html http/tests/media/hls/hls-video-resize.html http/tests/media/hls/video-controller-getStartDate.html http/tests/media/hls/video-controls-live-stream.html http/tests/media/hls/video-cookie.html http/tests/media/hls/video-duration-accessibility.html http/tests/security/anchor-download-allow-sameorigin.html Updating expectations accordingly. The HLS crashes seem to occur only on the release bot, which is strange. The other two occur on the debug bot too. Seems our bot crash catcher is still broken, there's no backtrace even on the debug bot, but I see this error message: STDERR: /home/slave/webkitgtk/gtk-linux-64-debug-tests/build/WebKitBuild/Debug/bin/NetworkProcess: No such file or directory. STDERR: ERROR: WebLoaderStrategy::networkProcessCrashed: failing all pending resource loaders STDERR: ../../Source/WebKit2/WebProcess/Network/WebLoaderStrategy.cpp(308) : void WebKit::WebLoaderStrategy::networkProcessCrashed() which (asides from the first line) is the same as what I see when I try creating an Epiphany web app, so I presume it's the same issue. The layout test crashes were introduced in r206731 "[SOUP] Cleanup persistent credential storage code" (In reply to comment #1) > Updating expectations accordingly. Actually I'm not going to update the expectations, because there are existing expectations for some of the HLS tests and it's annoying to try to assign expectations to multiple bugs, and because I am hoping we can fix this quickly. (In reply to comment #1) > which (asides from the first line) is the same as what I see when I try > creating an Epiphany web app, so I presume it's the same issue. This might be a bad assumption because any network process crash would cause the same message, but let's assume it's the same for now as hopefully we did not introduce multiple network process crashes around the same time. Created attachment 291100 [details]
Patch
Committed r206996: <http://trac.webkit.org/changeset/206996> |