#2554 closed defect (fixed)
Webstatus crash when viewing a builder
Reported by: | ludovicchabant | Owned by: | |
---|---|---|---|
Priority: | blocker | Milestone: | undecided |
Version: | 0.8.8 | Keywords: | |
Cc: |
Description
I frequently get this crash when I either want to view a builder's page, or a specific build's page:
web.Server Traceback (most recent call last): twisted.internet.defer.FirstError: FirstError[#0, [Failure instance: Traceback (failure with no frames): <class 'twisted.internet.defer.FirstError'>: FirstError[#0, [Failure instance: Traceback: <type 'exceptions.TypeError'>: 'NoneType' object has no attribute '__getitem__' C:\Python27\lib\site-packages\twisted\internet\defer.py:575:_runCallbacks D:\SnowMonkey\TnT\Build\Bin\SnowMonkey\buildbot\master\buildbot\util\lru.py:217:handle_result C:\Python27\lib\site-packages\twisted\internet\defer.py:380:callback C:\Python27\lib\site-packages\twisted\internet\defer.py:488:_startRunCallbacks --- <exception caught here> --- C:\Python27\lib\site-packages\twisted\internet\defer.py:575:_runCallbacks D:\SnowMonkey\TnT\Build\Bin\SnowMonkey\buildbot\master\buildbot\sourcestamp.py:120:<lambda> D:\SnowMonkey\TnT\Build\Bin\SnowMonkey\buildbot\master\buildbot\changes\changes.py:55:fromChdict ]] ]] twisted.internet.defer.FirstError: FirstError[#0, [Failure instance: Traceback (failure with no frames): <class 'twisted.internet.defer.FirstError'>: FirstError[#0, [Failure instance: Traceback: <type 'exceptions.TypeError'>: 'NoneType' object has no attribute '__getitem__' C:\Python27\lib\site-packages\twisted\internet\defer.py:575:_runCallbacks D:\SnowMonkey\TnT\Build\Bin\SnowMonkey\buildbot\master\buildbot\util\lru.py:217:handle_result C:\Python27\lib\site-packages\twisted\internet\defer.py:380:callback C:\Python27\lib\site-packages\twisted\internet\defer.py:488:_startRunCallbacks --- <exception caught here> --- C:\Python27\lib\site-packages\twisted\internet\defer.py:575:_runCallbacks D:\SnowMonkey\TnT\Build\Bin\SnowMonkey\buildbot\master\buildbot\sourcestamp.py:120:<lambda> D:\SnowMonkey\TnT\Build\Bin\SnowMonkey\buildbot\master\buildbot\changes\changes.py:55:fromChdict ]] ]]
It seems that "cache" in "None" in "Change.fromChdict".
The bug appears out of nowhere, i.e. one second it works, and the next it will fire all the time and I need to kill all the caches, along with the DB, and rebuild the master. Any idea what could be causing this? I could just quick-patch it and return "None" if "cache is None", but I'm not sure if that would be a proper fix or just delaying the problem.
Change History (3)
comment:1 Changed 8 years ago by dustin
comment:2 Changed 8 years ago by dustin
- Resolution set to fixed
- Status changed from new to closed
commit b4ffcdb1adeebff3bb601076903b3de1d78b7e37 Author: Dustin J. Mitchell <dustin@mozilla.com> Date: Sun Sep 1 10:52:52 2013 -0400 Omit changes that aren't in the DB from sourcestamps Fixes #2554.
should fix this up.
comment:3 Changed 8 years ago by ludovicchabant
Thanks! I'll report back if the bug shows up again... hopefully there won't be any reporting back :)
Note: See
TracTickets for help on using
tickets.
Not quite - chdict is None, meaning that the change with that ID is not in the DB. I don't know why that would be the case.
The right fix is to add code to ignore changes that aren't in the DB in sourcestamp.py.