1028119 - "TypeError: on_timeout() takes exactly 2 arguments (1 given)" in mozprocess/processhandler.py", line 724, in onTimeout

Status: RESOLVED FIXED

(Testing :: Mozbase, defect)

Description

[Ed Morley [:emorley]]

Broken out from bug 1027960.

Andrew, I think this is fallout from bug 997244, given the timing of the filing of bug 1027960 and that this has only been seen on B2G so far. Would you mind taking a look? :-)

Comment 1

[Ed Morley [:emorley]]

Sorry, the log:

https://tbpl.mozilla.org/php/getParsedLog.php?id=42134471&tree=B2g-Inbound
b2g_emulator_vm b2g-inbound opt test mochitest-4 on 2014-06-20 03:03:23 PDT for push 49f37e156352

{
04:01:22     INFO -  1148 INFO TEST-START | /tests/docshell/test/test_bug529119-1.html
04:01:24     INFO -  ###################################### forms.js loaded
04:01:24     INFO -  ############################### browserElementPanning.js loaded
04:01:24     INFO -  ######################## BrowserElementChildPreload.js loaded
04:01:27     INFO -  ############ ErrorPage.js
04:01:27     INFO -  ############ ErrorPage.js
04:01:29     INFO -  1149 INFO TEST-INFO | MEMORY STAT vsize after test: 131358720
04:01:29     INFO -  1150 INFO TEST-INFO | MEMORY STAT residentFast after test: 54571008
04:01:29     INFO -  1151 INFO TEST-INFO | MEMORY STAT heapAllocated after test: 27180984
04:07:00     INFO -  Exception in thread Thread-6:
04:07:00     INFO -  Traceback (most recent call last):
04:07:00     INFO -    File "/usr/lib/python2.7/threading.py", line 551, in __bootstrap_inner
04:07:00     INFO -      self.run()
04:07:00     INFO -    File "/usr/lib/python2.7/threading.py", line 504, in run
04:07:00     INFO -      self.__target(*self.__args, **self.__kwargs)
04:07:00     INFO -    File "/builds/slave/test/build/venv/local/lib/python2.7/site-packages/mozprocess/processhandler.py", line 786, in _processOutput
04:07:00     INFO -      self.onTimeout()
04:07:00     INFO -    File "/builds/slave/test/build/venv/local/lib/python2.7/site-packages/mozprocess/processhandler.py", line 724, in onTimeout
04:07:00     INFO -      handler()
04:07:00     INFO -  TypeError: on_timeout() takes exactly 2 arguments (1 given)
04:07:00     INFO -  Stopping web server
04:08:03     INFO -  Stopping web socket server
04:08:03     INFO -  Stopping ssltunnel
04:08:03     INFO -  Mochitest INFO | runtestsb2g.py | Running tests: end.
}

Comment 2

[Andrew Halberstadt [:ahal]]

Attachment: fix error on timeout

Bah, stupid mistake. It must have either not shown up in try, or I missed it when it did. Thanks for catching this!

Comment 3

[Andrew Halberstadt [:ahal]]

Comment on attachment 8443444 [details] [diff] [review]
fix error on timeout

Hang on, there's another error in there that this will expose..

Comment 4

[Andrew Halberstadt [:ahal]]

Attachment: fix error on timeout

This fixes a few other errors I noticed. Note this whole function should get moved somewhere else as part of bug 1025051. I may also need to get Ahmed to update his structured logging patch to do this in a structured way.

Comment 5

[Ed Morley [:emorley]]

Comment on attachment 8443451 [details] [diff] [review]
fix error on timeout

Review of attachment 8443451 [details] [diff] [review]:
-----------------------------------------------------------------

lgtm; thank you for the quick turnaround :-)

Comment 6

[Andrew Halberstadt [:ahal]]

https://hg.mozilla.org/integration/mozilla-inbound/rev/4cffdbeec391

Comment 7

[Ed Morley [:emorley]]

This somehow has turned debug mochitest-9 orange?
https://tbpl.mozilla.org/?tree=Mozilla-Inbound&jobname=mochitest-debug-9&tochange=da007249e737&fromchange=3bf0b88186bf

Comment 8

[Andrew Halberstadt [:ahal]]

Attachment: change timeout string

This should hide the perma-orange in m-9, but not fix it. We still need to investigate why this is happening, and why it wasn't previously turning orange. There is a chance that b2g emulator timeouts haven't been working properly for awhile now.

Comment 9

[Ed Morley [:emorley]]

Comment on attachment 8443597 [details] [diff] [review]
change timeout string

I'm only partially sure we may trick the parser this way - it may be that the refactor fixed something else too. Worth a try though (failing this, I guess we just hide debug m-9 until the shutdown timeout is fixed).

Comment 10

[Andrew Halberstadt [:ahal]]

Yeah, that could be. I guess we'll find out soon:
https://hg.mozilla.org/integration/mozilla-inbound/rev/8965c073da53

Comment 11

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/4cffdbeec391
https://hg.mozilla.org/mozilla-central/rev/8965c073da53