httpd-bugs mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From bugzi...@apache.org
Subject DO NOT REPLY [Bug 44402] - Worker mpm crashes (SEGV) under stress with static workload on solaris x86
Date Sat, 16 Feb 2008 08:25:04 GMT
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG·
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
<http://issues.apache.org/bugzilla/show_bug.cgi?id=44402>.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND·
INSERTED IN THE BUG DATABASE.

http://issues.apache.org/bugzilla/show_bug.cgi?id=44402





------- Additional Comments From rpluem@apache.org  2008-02-16 00:25 -------
(In reply to comment #12)

> If you agree that there is a clear race condition then to correct this, I have
> following suggestion :
> (a) Use a dedicated pool for each worker thread, this will avoid any locking.
> It will perform better but may require little more memory in those situations
> when worker threads are not fully used.
> (b) Use some other technique other than a recycled pool list which avoids race
> conditions.
> 
> I am in favour of option (a) until some good idea for (b) comes to my mind.
> If you agree with (a) then I can work and generate a patch.
> 
> Note : Also I believe that the crash will happen in linux too. I never ran more
> than 1 hour in linux. I will try that tonight.
> 

Thank you for your thorough investigation. I agree with you that we have the
described race conditions here. We have a similar race in the event MPM.
Next steps:

1. Bring your patch above into trunk. Currently I see no significant performance 
   loss over the current code as we are using a mutex there as well. We only
   increase the time during which we lock the resource. I don't know right
   now when I find the cycles to apply the patch to trunk, but if you could
   attach a trunk version of your patch to this report it would be a big help.

2. Move the further discussion regarding options a) or b) to 
   dev@httpd.apache.org and lets wait for its results to decide how to move 
   along and improve the situation here in the long run.
 


-- 
Configure bugmail: http://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org


Mime
View raw message