[Webkit-unassigned] [Bug 39198] New: Make sheriffbot error with human readable message instead of backtrace when rollout fails
bugzilla-daemon at webkit.org
bugzilla-daemon at webkit.org
Mon May 17 00:24:23 PDT 2010
https://bugs.webkit.org/show_bug.cgi?id=39198
Summary: Make sheriffbot error with human readable message
instead of backtrace when rollout fails
Product: WebKit
Version: 528+ (Nightly build)
Platform: All
OS/Version: All
Status: NEW
Severity: Normal
Priority: P2
Component: Tools / Tests
AssignedTo: webkit-unassigned at lists.webkit.org
ReportedBy: dbates at webkit.org
CC: eric at webkit.org, hyatt at apple.com, abarth at webkit.org
We should make sheriffbot error with a human readable message instead of a python backtrace when it fails to rollout a patch.
When David Hyatt tried to rollout r59274 tonight (05/17), sheriffbot died with the following message:
sheriffbot: Failed to run "['./WebKitTools/Scripts/webkit-patch', '--status-host=webkit-commit-queue.appspot.com', 'create-rollout', '--force-clean', '--parent-command=sheriff-bot', 59274, 'Not worth the memory hit. Just roll this out. (Requested by dhyatt on #webkit).']" exit_code: 2
to which Hyatt responded:
[8:43pm] dhyatt: exit_code: 2!
Clearly, "exit_code: 2" is not very informative. After further inspection, it appears that the reason for the error was that r59274 could not be rolled out cleanly. That is, there were conflicts with some newer changes.
Adam Barth helped clarify the meaning of the various error_codes:
[9:31pm] abarth: we use three different exit codes (if you care)
[9:31pm] abarth: 0 = success
[9:31pm] abarth: 1 = failure and I dealt with it
[9:31pm] abarth: 2 = failure and you need to deal with it
[9:31pm] abarth: "you" being the master process
[9:31pm] abarth: (or maybe 1 and 2 are reversed, i don't remember)
--
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.
More information about the webkit-unassigned
mailing list