codebrowse is hanging/unresponsive

Bug #1015716 reported by Matthew Wedgwood
16
This bug affects 3 people
Affects Status Importance Assigned to Milestone
Launchpad itself
Triaged
Critical
Unassigned

Bug Description

We're seeing codebrowse get hung up on the same project over and over. Process becomes intermittently unresponsive.

Sample thread dumps from a recent restart:
https://pastebin.canonical.com/68589/
https://pastebin.canonical.com/68590/

Core dumps:
https://pastebin.canonical.com/68591/
https://pastebin.canonical.com/68592/

Revision history for this message
Gary Poster (gary) wrote :

thedac said at the time: "https://pastebin.canonical.com/68586/ we get these longer running branches but then they seem to go away"

Changed in launchpad:
status: New → Triaged
importance: Undecided → Critical
tags: added: codebrowse
Revision history for this message
Matthew Wedgwood (mew) wrote :

I may be mistaken on the "same project" part. Some of the long-running threads from the thread dumps are from our nagios checks and the code doesn't change. Still investigating.

Revision history for this message
John A Meinel (jameinel) wrote :

Note, I'm not able to get a page to load for the gcc tree, I'm guessing this is just 'loggerhead is slow' and it is slow enough that it is unable to preload its history cache for such a large branch. (namely this url always fails to load, but I'm unable to get any alternate URL of that branch to load either: http://bazaar.launchpad.net/%7Elinaro-toolchain-dev/gcc-linaro/4.5/revision/99503/gcc/testsuite/gcc.target/arm/neon/vst2_laneu8.c )

Revision history for this message
David Ames (thedac) wrote :
Revision history for this message
David Ames (thedac) wrote :

Core dump from when loggerhead was hung on carob:~dames/core.31162.bz2

Revision history for this message
Robert Collins (lifeless) wrote : Re: [Bug 1015716] Re: codebrowse is hanging/unresponsive

So, this reflects:
 - performance and reliability issue within codebrowse.

We need to address it. The proximate triggering of it was a user
mirroring the codebrowse content as far as we can tell, which isn't an
intended use-case - they have been blocked to give us breathing spac.e

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Duplicates of this bug

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.