Bug 78 - Implement maximum number of children
Summary: Implement maximum number of children
Status: RESOLVED FIXED
Alias: None
Product: Spamassassin
Classification: Unclassified
Component: spamc/spamd (show other bugs)
Version: 2.30CVS
Hardware: All All
: P2 enhancement
Target Milestone: ---
Assignee: Craig Hughes
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2002-03-03 18:16 UTC by Duncan Findlay
Modified: 2002-06-15 03:58 UTC (History)
0 users



Attachment Type Modified Status Actions Submitter/CLA Status
Patch implementing maximum number of children patch None Duncan Findlay [HasCLA]

Note You need to log in before you can comment on or make changes to this bug.
Description Duncan Findlay 2002-03-03 18:16:49 UTC
My computer is too slow, OR, spamd uses way too much memory on my P1/100 40MB 
RAM.

It would be great to limit the number of children forked by spamd.
Comment 1 Duncan Findlay 2002-03-03 18:20:41 UTC
Created attachment 3 [details]
Patch implementing maximum number of children
Comment 2 Duncan Findlay 2002-03-03 18:21:53 UTC
I'm only a bit concerned that this might break BSD with the -m option.
However, I believe BSD queues only 5 connections, so it would be unwise to use 
this option in the first place.
Comment 3 Craig Hughes 2002-03-04 18:19:53 UTC
Pending mailing list discussion of sighandler
Comment 4 Craig Hughes 2002-03-17 04:47:16 UTC
I think the concensus is the patch breaks *BSD compatibility -- did you hear any
suggestions on how to make it better?  If not, I think I'll just leave this bug
in the DB here and we can refer people to it if they want to limit the number of
child processes.
Comment 5 Duncan Findlay 2002-03-24 14:41:08 UTC
I disagree.
There has been no consensus. I'm not convinced anyone has even tested the patch.
Read the thread 'Request for testing on BSD / Limiting children for spamd' again.

Furthermore, the patch leaves everything the same if the -m option is not given.

I'd really like this to be implemented. When I came back from vacation yesterday, all 397 messages from fetchmail were processed without slowing my machine to the point where I couldn't even log in. (as it would have without this)
Comment 6 Craig Hughes 2002-03-25 17:20:18 UTC
Ok, well I gave the BSDers a comment period :)  If this does cause BSD 
problems, we'll fix it.  For now I'm adding the patch to CVS.