[run_web_tests.py] Keep stderr of the last failure for repeated tests
This is useful for debugging flaky tests with --iterations=n (or repeat-each=n) and --exit-after-n-failures=1. Previously the first stderr was kept which might be for a pass which was confusing in results.html where the failure results and the passing stderr were displayed. This doesn't affect normal runs (e.g. on bots) where tests are not repeated in one run (not including retries). Also output stderr for timeout tests, which is also useful for debugging. Change-Id: I25c1296e14b22763f36eb96d4028896f28a11c36 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2284215Reviewed-by:Robert Ma <robertma@chromium.org> Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org> Cr-Commit-Position: refs/heads/master@{#785827}
Showing
Please register or sign in to comment