I'm running several QMT servers with spamdyke, and am now of the opinion that there's a bug in spamdyke. All are running v4.0.10. Also, I believe that DNS resolution is configured and working properly in all cases (local caching DNS, forwarding to root servers), so that should not be an issue.
On low volume servers, only 1 of 3 had any defunct processes at all. They were few (1-2 per month) and very old (1-2 months old). Where I still had the corresponding log entries, I saw the tcpserver:pid and tcpserver:ok messages, but nothing else. No further messages from any smtp-related program. On a high volume server, defunct processes are much more frequent. They all appear to be sessions with a spamdyke:TIMEOUT message, although there are also many TIMEOUTs which do not result in defunct processes. The defunct sessions vary as to the type of rejection, some rDNS, some RBLs, but they all eventually get a TIMEOUT message, but no subsequent tcpserver:end message. I'm going to see if I can get a detail log from a session that goes defunct. Sam, if there's anything else I can do to help pinpoint this thing, let me know. -- -Eric 'shubes' Sam Clippinger wrote: > If this happens again, would you mind using strace to see where spamdyke > is stuck? It would be very helpful to know as much as possible so I can > try to figure out what's happening. > > -- Sam Clipipnger > > On 2/26/10 1:23 AM, Andreas Galatis wrote: >> Hi Hans, >> I did not receive your answer, just got it from the archive now. >> >> you're right, the server was unstable/slow and i had hanging SpamDyke >> processes. >> Since DNS-resolver is ok, I have a stable server and no hanging processes. >> Shure, SpamDyke should end processes, even when the resolver doesn't respond >> in time, but both problems where solved after reconfiguring DNS-cache. >> >> Andreas >> >> >>> I assuming that with "the same problem" you are referring to a >>> slow/unstable server, and not the hanging SpamDyke processes? >>> No matter what the problem is, I don't think there should be >>> SpamDyke processes hanging around. >>> >>> Hans >>> >> >>> Hi Hans, >>> >>> I had the same problem in the past and ended up that my real problem was >>> the dns-resolver. >>> With a working dnscache all my problems with where gone. >>> Jm2c >>> >>> Andreas >>> >>> Am Thursday 25 February 2010 11:47:01 schrieb Hans F. Nordhaug: >>> >>>> Hi! >>>> >>>> Today we turned of Spamdyke to see if it makes our e-mail server more >>>> stable. The server is running a plain, up-to-date CentOS 5.3 with >>>> SpamDyke 4.0.10 and Qmail from Qmailtoaster/Qmailtoaster Plus. >>>> >>>> What we are seeing is 100+ hanging Spamdyke processing and >>>> corresponding defunct qmail-smtpd child processes. We haven't >>>> monitored the number of hanging processes, but when we get reports >>>> from our users about problems with slow e-mail sending/receiving that >>>> is what we find. Killing these hanging Spamdyke processes makes the >>>> e-mail speed-up. We do monitor the actual Qmal local/remote queue and >>>> it is less than 20 so the problem seems to be the connections in some >>>> way. It's all very weird because we shouldn't be near the max number >>>> of sockets, right? It might all be coincidences, but we have >>>> experienced the problem (and the fix) three times so far in 2010. >>>> >>>> Anyway, why aren't these Spamdyke processes stopping. (The time-out is >>>> 180 seconds and these processes are many hours/days old.) >>>> >>>> Thanks for any input. >>>> >>>> Regards, >>>> Hans >>>> >>>> PS! I might turn Spamdyke on again for a specific domain (and/or port) if >>>> you want to test it or want me to run some tests. _______________________________________________ spamdyke-users mailing list spamdyke-users@spamdyke.org http://www.spamdyke.org/mailman/listinfo/spamdyke-users
