(So it's clearer what the processes still alive after shutdown were spawned for / by what...)

This gave output of the form: { 04:04:20 INFO - INFO | automation.py | Application ran for: 0:33:08.077000 04:04:20 INFO - INFO | automation.py | Reading PID log: c:\users\cltbld\appdata\local\temp\tmp3icbonpidlog 04:04:20 INFO - ==> process 624 launched child process 2980 (") 04:04:20 INFO - ==> process 624 launched child process 3572 (") 04:04:20 INFO - ==> process 624 launched child process 1172 (") 04:04:20 INFO - ==> process 624 launched child process 3512 (") 04:04:20 INFO - ==> process 624 launched child process 2664 (") ... } https://tbpl.mozilla.org/?tree=Try&rev=c9d5ccae9936 Not sure what I'm missing? Do I need to cast |cmdline.c_str()|, or is it actually the empty string?

(In reply to Ed Morley [:edmorley UTC+1] from comment #1) > Not sure what I'm missing? Do I need to cast |cmdline.c_str()|, or is it > actually the empty string?

You need %S there, not %s, since it's a UTF-16 string (wstring, wchar_t*).

Thank you for that :-) https://tbpl.mozilla.org/?tree=Try&rev=ad2e6205c66c

Any progress here?

(In reply to ben turner [:bent] from comment #5) > Any progress here? The previous Try run results have expired, rebased + resubmitted: https://tbpl.mozilla.org/?tree=Try&rev=71ec65aa72f8

Example log (sadly not yet got it to fail on Try): https://tbpl.mozilla.org/php/getParsedLog.php?id=25023330&tree=Try { ... 10:24:59 INFO - ==> process 3208 launched child process 1116 ("C:\slave\test\build\application\firefox\plugin-container.exe" --channel=3208.7d96e30.197057342 -greomni "C:\slave\test\build\application\firefox\omni.ja" -appomni "C:\slave\test\build\application\firefox\browser\omni.ja" -appdir "C:\slave\test\build\application\firefox\browser" - 3208 "\\.\pipe\gecko-crash-server-pipe.3208" tab) 10:24:59 INFO - ==> process 3208 launched child process 2744 ("C:\slave\test\build\application\firefox\plugin-container.exe" --channel=3208.7d97d30.1275298728 -greomni "C:\slave\test\build\application\firefox\omni.ja" -appomni "C:\slave\test\build\application\firefox\browser\omni.ja" -appdir "C:\slave\test\build\application\firefox\browser" - 3208 "\\.\pipe\gecko-crash-server-pipe.3208" tab) 10:24:59 INFO - ==> process 3208 launched child process 1548 ("C:\slave\test\build\application\firefox\plugin-container.exe" --channel=3208.7d96e30.503169550 -greomni "C:\slave\test\build\application\firefox\omni.ja" -appomni "C:\slave\test\build\application\firefox\browser\omni.ja" -appdir "C:\slave\test\build\application\firefox\browser" - 3208 "\\.\pipe\gecko-crash-server-pipe.3208" tab) 10:24:59 INFO - ==> process 3208 launched child process 1604 ("C:\slave\test\build\application\firefox\plugin-container.exe" --channel=3208.edb8120.847868485 -greomni "C:\slave\test\build\application\firefox\omni.ja" -appomni "C:\slave\test\build\application\firefox\browser\omni.ja" -appdir "C:\slave\test\build\application\firefox\browser" - 3208 "\\.\pipe\gecko-crash-server-pipe.3208" tab) ... }

Thank you :-) https://hg.mozilla.org/integration/mozilla-inbound/rev/fbe4ea122205

Funny enough, this patch apparently made bug 603147 disappear. Confirmed via Try bisection and by applying the patch on top of mozilla-beta (Fx24) and retriggering like crazy. https://tbpl.mozilla.org/?tree=Try&rev=c09dc4901ffd Can we please uplift this for the ESR?

Comment on attachment 735649 [details] [diff] [review] Patch v1 [Approval Request Comment] Bug caused by (feature/regressing bug #): n/a User impact if declined: sheriff pain for the next year on esr24 Testing completed (on m-c, etc.): been on m-c for since early July Risk to taking this patch (and alternatives if risky): minimal, it's a trivial logging change String or IDL/UUID changes made by this patch: none

Assuming no QA needed here. Please remove [qa-] from the whiteboard and add the verifyme keyword if this needs QA.