Closed Bug 1297575 Opened 6 years ago Closed 6 years ago
.html/test/browser/browser _toolbox _swap _browsers .js | This test exceeded the timeout threshold . It should be rewritten or split up . If that's not possible, use request Longer Timeout(N), but only as a last resort . -
Priority: -- → P3
Practically all failures are on linux64-asan, e10s. This e10s-only test runs in < 10 seconds on linux64 opt/pgo but typically takes > 40 seconds on linux64-asan. At 45 seconds, it times out.
It feels like requestLongerTimeout() would be appropriate for this test on asan...but not so much for other platforms. Looking for suggestions for how to deal with this, and similar cases of run-too-long-on-platform (maybe bug 1292233). Some possible approaches: - requestLongerTimeout() - skip-if = asan # runs for too long on asan - requestLongerTimeout() for this test, on asan only, somehow - increase default timeout on asan for all mochitests, somehow
I like the list of approaches here- I would question if running this test on ASAN brings value and if it isn't important we could do less ASAN testing :) Assuming it is important to run this test on ASAN, I think option 3 or 4. The simplest thing would be #4, but that would be harder to debug/understand what is going on- which leads me to #3. How do we do that, I assume we could/should pass in mozinfo data to the browser and requestLogerTimeout could have the data from mozinfo :)
As with bug 1292233, I don't think this test itself can be made much faster... I would be okay with any of the approaches in comment 11.
I'm proceeding with a variation on the last option in Comment 11, increasing the default timeout on asan for all 'browser' flavored mochitests, from 45 to 90 seconds - bug 1313187. If that works, I'll dup this bug to that one.
Assignee: nobody → gbrown
Status: NEW → RESOLVED
Closed: 6 years ago
Resolution: --- → DUPLICATE
Duplicate of bug: 1313187
Bug 1313187 resolved all the Linux64-asan instances (the majority), as expected. But the test also runs too long on win32-debug occasionally.
Status: RESOLVED → REOPENED
Resolution: DUPLICATE → ---
Let's requestLongerTimeout to fix the rest.
Assignee: gbrown → jryans
Status: REOPENED → ASSIGNED
Priority: P3 → P1
Pushed by email@example.com: https://hg.mozilla.org/integration/mozilla-inbound/rev/66a58018adea Longer timeout for browser_toolbox_swap_browsers.js. r=me
You need to log in before you can comment on or make changes to this bug.