Bisect - Tee'ing output seems to break Windows |
|||||||||
Issue descriptionhttps://bugs.chromium.org/p/chromium/issues/detail?id=683680#c3 Traceback (most recent call last): File "C:\b\rr\tmp9hcs2v\rw\checkout\scripts\slave\tee.py", line 41, in <module> sys.exit(main()) File "C:\b\rr\tmp9hcs2v\rw\checkout\scripts\slave\tee.py", line 23, in main pipe = subprocess.Popen(cmd, stdout=subprocess.PIPE) File "C:\b\depot_tools\python276_bin\lib\subprocess.py", line 709, in __init__ errread, errwrite) File "C:\b\depot_tools\python276_bin\lib\subprocess.py", line 957, in _execute_child startupinfo) WindowsError: [Error 193] %1 is not a valid Win32 application
,
Jan 23 2017
,
Jan 23 2017
,
Jan 23 2017
,
Jan 23 2017
Upping priority, looks like Win bisects in general are broken.
,
Jan 23 2017
,
Jan 23 2017
Landed this, https://chromium-review.googlesource.com/c/431066/ which potentially fixes this issue. Will keep an eye on the Windows bisect bots.
,
Jan 23 2017
Seems like the Windows bisect builder bot is broken which is preventing me from verifying if my fix worked. Builder failure seems unrelated to my changes. But who knows. https://uberchromegw.corp.google.com/i/tryserver.chromium.perf/builders/winx64_bisect_builder/builds/5049
,
Jan 23 2017
,
Jan 23 2017
Filed a bug on the broken windows builder.
,
Jan 24 2017
The fix seemed to have worked for tee-ing the output. For example, see: https://luci-logdog.appspot.com/v/?s=chrome%2Fbb%2Ftryserver.chromium.perf%2Fwinx64_10_perf_bisect%2F854%2F%2B%2Frecipes%2Fsteps%2FGathering_reference_values%2F0%2Fsteps%2FPerformance_Test_3_of_6%2F0%2Fstdout Test appears to run normally wrapped by tee.py script.
,
Jan 24 2017
Awesome, thank you!
,
Jan 24 2017
Thank you Simon and Mike.
,
Feb 3 2017
|
|||||||||
►
Sign in to add a comment |
|||||||||
Comment 1 by simonhatch@chromium.org
, Jan 23 2017