[chromium-dev] Re: try failure for 'pinkerton: appmode' try job on win @ r36656

2010-01-21 Thread Marc-Antoine Ruel
[+chromium-dev and sheriffs] These are flaky tests. It is disturbing since there is a crash in a really normal UI workflow (bookmarks) Whoever knows about these tests should work on fixing these or putting these as flaky (except the last one since it's a crasher): -

Re: [chromium-dev] Re: try failure for 'pinkerton: appmode' try job on win @ r36656

2010-01-21 Thread Thomas Van Lenten
DownloadFilename is a strings error, the bot needs a clobber. We had this on the main waterfall yesterday morning also and the few that saw it got fixed by a clobber. TVL On Thu, Jan 21, 2010 at 9:36 AM, Marc-Antoine Ruel mar...@chromium.orgwrote: [+chromium-dev and sheriffs] These are

Re: [chromium-dev] Re: try failure for 'pinkerton: appmode' try job on win @ r36656

2010-01-21 Thread Marc-Antoine Ruel
Clobber Done. On Thu, Jan 21, 2010 at 9:44 AM, Thomas Van Lenten thoma...@chromium.orgwrote: DownloadFilename is a strings error, the bot needs a clobber. We had this on the main waterfall yesterday morning also and the few that saw it got fixed by a clobber. TVL On Thu, Jan 21, 2010 at

Re: [chromium-dev] Re: try failure for 'pinkerton: appmode' try job on win @ r36656

2010-01-21 Thread Marc-Antoine Ruel
With regards to BookmarkContextMenuTest.DeleteURL, I filed http://crbug.com/32784. Glen, you should make sure this bug is triaged accordingly since flaky crashers can't be tolerated. M-A On Thu, Jan 21, 2010 at 9:45 AM, Marc-Antoine Ruel mar...@chromium.orgwrote: Clobber Done. On Thu, Jan

Re: [chromium-dev] Re: try failure for 'pinkerton: appmode' try job on win @ r36656

2010-01-21 Thread Scott Violet
BookmarkContextMenuTest.DeleteURL is not listed on the flakiness dashboard. BookmarkContextMenuTest relies indirectly on string ids as well. The clobber should have fixed it. Is there no way we can make the test harnesses fail early on if the ids change but everything isn't recompiled? -Scott