Summary: | Assertion failure in PlatformPasteboardMac constructor causing multiple tests to "crash" on the Lion Intel WebKit2 Debug bots | ||
---|---|---|---|
Product: | WebKit | Reporter: | Jessie Berlin <jberlin> |
Component: | HTML Editing | Assignee: | Nobody <webkit-unassigned> |
Status: | NEW --- | ||
Severity: | Normal | CC: | aakash_jain, ap, enrica, jberlin, rniwa, thorton, webkit-bug-importer |
Priority: | P2 | Keywords: | InRadar, LayoutTestFailure, Regression |
Version: | 528+ (Nightly build) | ||
Hardware: | Mac (Intel) | ||
OS: | OS X 10.7 | ||
URL: | http://build.webkit.org/results/Lion%20Intel%20Debug%20(WebKit2%20Tests)/r110566%20(4846)/editing/execCommand/copy-without-selection-crash-log.txt |
Description
Jessie Berlin
2012-03-13 11:03:41 PDT
Killing the pasteboard server seemed to fix the crash on at least one of the Mac Lion Debug WK2 testers. Enrica and I were unable to reproduce these crashes locally. We think that some other tests crashing might be leaving the system / pasteboard in a weird state. Hopefully we can fix more of the other test crashes soon so we don't get into this state again. I am going to kill the pasteboard server on all the affected bots. If we see this crash come back, it might be worth adding a build step to kill the pasteboard server. Added a step to kill the pasteboard server in http://trac.webkit.org/changeset/111866. Hopefully this will remove the need for us to continually ssh into the affected bots and kill it manually until we can figure out what the underlying issue (and fix) is. Does this bug still exist? (In reply to comment #4) > Does this bug still exist? Until someone has the time to take out the kill step added in http://trac.webkit.org/changeset/111866 and actually debug the issue, we need to keep this bug open. AFAIK, nothing has changed (other than adding the kill step) that should have caused the issue to disappear. reverted this in https://bugs.webkit.org/show_bug.cgi?id=152426 |