Bisects running 0 tests |
|||||||||||
Issue descriptionI'm going to follow up and block a bunch of bugs on this one, but filing with a few examples. I believe this is likely related to the recent --story-filter work; cc-ing Juan, Charlie, Ned, and Ethan who may have investigated pieces of this already. https://build.chromium.org/p/tryserver.chromium.perf/builders/winx64intel_perf_bisect/builds/1223 Expand "Gathering Reference Values" step Click on any json.output from "Gathering reference values.Reading chartjson results" step and see that there are no values: https://build.chromium.org/p/tryserver.chromium.perf/builders/winx64intel_perf_bisect/builds/1223/steps/Gathering%20reference%20values.Reading%20chartjson%20results%20%2810%29/logs/json.output Click on any "Captured Output" from any "Gathering reference values.Performance Test" step and see that no tests ran: "[ PASSED ] 0 tests." https://build.chromium.org/p/tryserver.chromium.perf/builders/winx64intel_perf_bisect/builds/1223/steps/Gathering%20reference%20values.Performance%20Test%201%20of%205/logs/Captured%20Output Many bisects are failing due to this. Can anyone take a look?
,
Nov 4 2016
Some example command lines: https://build.chromium.org/p/tryserver.chromium.perf/builders/winx64ati_perf_bisect/builds/1638 C:\b\c\b\winx64ati_perf_bisect\src\tools\perf\run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter='http...msn\.co\.il.' page_cycler_v2.intl_ar_fa_he --output-dir=c:\users\chrome~1\appdata\local\temp\perf-test-outputskrp0u https://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus6_perf_bisect/builds/2725 /b/c/b/android_nexus6_perf_bisect/src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter='http...ynet\.co\.il.' page_cycler_v2.intl_ar_fa_he --device ZX1G229B6Q --output-dir=/tmp/perf-test-outputfv7HcL https://build.chromium.org/p/tryserver.chromium.perf/builders/linux_perf_bisect/builds/6818 /usr/bin/python /b/c/b/linux_perf_bisect/src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter='http...bookish\.com' page_cycler_v2.tough_layout_cases --output-dir=/tmp/perf-test-outputFjgEHu https://uberchromegw.corp.google.com/i/internal.tryserver.clankium/builders/clankium-tablet-perf-bisect/builds/9 /usr/bin/python /b/build/slave/clankium-tablet-perf-bisect/build/src/tools/perf/run_benchmark -v --browser=android-chrome --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter='http...kapook\.com' start_with_url.warm.startup_pages --device 09e26f10 --output-dir=/tmp/perf-test-outputJuWG3l https://build.chromium.org/p/tryserver.chromium.perf/builders/winx64intel_perf_bisect/builds/1223 C:\b\depot_tools\python276_bin\python.exe C:\b\c\b\winx64intel_perf_bisect\src\tools\perf\run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter='http...us\.24h\.com\.vn.' page_cycler_v2.intl_ko_th_vi --output-dir=c:\users\chrome~1\appdata\local\temp\perf-test-outputgdppkd https://build.chromium.org/p/tryserver.chromium.perf/builders/winx64ati_perf_bisect/builds/1639 C:\b\depot_tools\python276_bin\python.exe C:\b\rr\tmpheiz9e\rw\checkout\scripts\slave\runisolatedtest.py --test_name "" --builder_name winx64ati_perf_bisect --checkout_dir C:\b\c\b\winx64ati_perf_bisect C:\b\c\b\winx64ati_perf_bisect\src\tools\perf\run_benchmark -- C:\b\depot_tools\python276_bin\python.exe C:\b\c\b\winx64ati_perf_bisect\src\tools\perf\run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter='http...ru\.wikipedia\.org.' page_cycler_v2.intl_hi_ru --output-dir=c:\users\chrome~1\appdata\local\temp\perf-test-outputuclv5_
,
Nov 4 2016
Looks like there is something funny with the way the command line is interepreted in this one: https://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus9_perf_bisect/builds/2234/steps/Gathering%20reference%20values.Performance%20Test%201%20of%205/logs/Failure%20Output /usr/bin/python /b/c/b/android_nexus9_perf_bisect/src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter='Wikipedia\ \(1\ tab\)' smoothness.sync_scroll.key_mobile_sites_smooth --device HT4CTJT03478 --output-dir=/tmp/perf-test-outputAapHgb No benchmark named "\(1\".
,
Nov 7 2016
That's interesting. I'll take this to investigate further what is going on.
,
Nov 7 2016
Thanks, Juan!
,
Nov 7 2016
Catapult side bug: https://github.com/catapult-project/catapult/issues/2967
,
Nov 7 2016
A short term fix is in place: https://codereview.chromium.org/2481773002/ I've re-started some bisects on all of the blocked bugs.
,
Nov 8 2016
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/cfa43000d1547324bc97b8c1bf6a1b9303e6e8fb commit cfa43000d1547324bc97b8c1bf6a1b9303e6e8fb Author: catapult-deps-roller <catapult-deps-roller@chromium.org> Date: Tue Nov 08 00:10:04 2016 Roll src/third_party/catapult/ 8ac279fd8..d18e525a0 (6 commits). https://chromium.googlesource.com/external/github.com/catapult-project/catapult.git/+log/8ac279fd8db3..d18e525a0681 $ git log 8ac279fd8..d18e525a0 --date=short --no-merges --format='%ad %ae %s' 2016-11-07 aiolos Revert of [Telemetry] Remove the logic of killing stray adb processes (patchset #1 id:1 of https://codereview.chromium.org/2424843002/ ) 2016-11-07 aiolos Add missing import in model track. 2016-11-07 jbudorick [devil + systrace] Set up logging in profile_chrome. 2016-11-07 jbudorick [devil] Add explicit None checking to device_temp_file. 2016-11-07 perezju [Dashboard] Fix regex for complex story names 2016-11-07 nednguyen [Telemetry] Remove the logic of killing stray adb processes BUG=662301, 662472 , 656224 Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+/master/autoroll/README.md If the roll is causing failures, see: http://www.chromium.org/developers/tree-sheriffs/sheriff-details-chromium#TOC-Failures-due-to-DEPS-rolls CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.android:android_optional_gpu_tests_rel TBR=catapult-sheriff@chromium.org Review-Url: https://codereview.chromium.org/2477333004 Cr-Commit-Position: refs/heads/master@{#430433} [modify] https://crrev.com/cfa43000d1547324bc97b8c1bf6a1b9303e6e8fb/DEPS
,
Nov 8 2016
Lowering pri- here as the workaround is now in place. Also have a CL ready for one of the places where the command line was being interpreted incorrectly.
,
Nov 9 2016
The following revision refers to this bug: https://chromium.googlesource.com/chromium/tools/build.git/+/870e91d32948d0c30e4bb5833e6088a3c2a50a23 commit 870e91d32948d0c30e4bb5833e6088a3c2a50a23 Author: Juan A. Navarro Perez <perezju@chromium.org> Date: Wed Nov 09 10:36:58 2016 [Bisect Script] Correctly interpret quoted arguments in test command Allow test commands to use shell-like quoting for passing arguments with special characters (e.g. spaces, etc.) BUG= 662472 Change-Id: Ib2794bcce1dba8f5dd195f9defddd3dc3a16a2fc Reviewed-on: https://chromium-review.googlesource.com/408139 Commit-Queue: Aaron Gable <agable@chromium.org> Reviewed-by: Aaron Gable <agable@chromium.org> [modify] https://crrev.com/870e91d32948d0c30e4bb5833e6088a3c2a50a23/scripts/slave/recipe_modules/bisect_tester/perf_test.py [modify] https://crrev.com/870e91d32948d0c30e4bb5833e6088a3c2a50a23/scripts/slave/recipe_modules/bisect_tester/perf_test_unittest.py
,
Nov 10 2016
,
Nov 10 2016
,
Nov 10 2016
,
Nov 16 2016
,
Nov 18 2016
Okay, so the priority has been lowered, but what are the next steps here? If there's a better long-term fix, maybe that should be filed as a fresh bug and this one should be closed?
,
Nov 25 2016
Yeah, I agree with closing this for now, the current fixes in place have served us well. The longer term fix would fit within the realm of the data pipe from benchmarks to dashboard.
,
Feb 3 2017
|
|||||||||||
►
Sign in to add a comment |
|||||||||||
Comment 1 by nedngu...@google.com
, Nov 4 2016