Bug 84611
| Summary: | EWS bots have no timeout protection for the c++ unittests | ||
|---|---|---|---|
| Product: | WebKit | Reporter: | Eric Seidel (no email) <eric> |
| Component: | Tools / Tests | Assignee: | Nobody <webkit-unassigned> |
| Status: | NEW | ||
| Severity: | Normal | CC: | abarth, ap, eric, gavinp, jake.nielsen.webkit, jamesr, spena |
| Priority: | P2 | ||
| Version: | 528+ (Nightly build) | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
Eric Seidel (no email)
All Chromium EWS bots are hung running unit tests
I guess the c++ unittests have no hang protection.
[----------] 8 tests from TreeSynchronizerTest
[ RUN ] TreeSynchronizerTest.syncNullTree
[ OK ] TreeSynchronizerTest.syncNullTree (0 ms)
[ RUN ] TreeSynchronizerTest.syncSimpleTreeFromEmpty
[ OK ] TreeSynchronizerTest.syncSimpleTreeFromEmpty (0 ms)
[ RUN ] TreeSynchronizerTest.syncSimpleTreeReusingLayers
(hang here forever)
Some patch 1.5 days ago hung all the bots. Still figuring out which patch. :)
| Attachments | ||
|---|---|---|
| Add attachment proposed patch, testcase, etc. |
Eric Seidel (no email)
cr-linux #1 is hung processing bug 84524.
Eric Seidel (no email)
Same hang for cr-linux-3:
[----------] 8 tests from TreeSynchronizerTest
[ RUN ] TreeSynchronizerTest.syncNullTree
[ OK ] TreeSynchronizerTest.syncNullTree (0 ms)
[ RUN ] TreeSynchronizerTest.syncSimpleTreeFromEmpty
[ OK ] TreeSynchronizerTest.syncSimpleTreeFromEmpty (0 ms)
[ RUN ] TreeSynchronizerTest.syncSimpleTreeReusingLayers
Was also hung on bug 84524
James Robinson
The googletest harness doesn't have appear to have a built-in test timeout mechanism: http://code.google.com/p/googletest/issues/detail?id=348, so this would probably have to be written in the python layer. We run the unit tests in a child process so that probably wouldn't be too hard.