It looks like this may be a pperl issue, the problems are occurring before it gets to my $qpsmtpd = Qpsmtpd::TcpServer->new();
So if i drop pperl and lower the concurrency to 50, it works ok, but 0.04 still seems to be processing faster but still hitting limits that if i ran 0.04 with the mods wouldnt happen. I dropped the -p flag as well for tcpserver to check but that seems to make no difference. ----- Original Message ----- From: "ross mueller" <[EMAIL PROTECTED]> To: "Ask Bjoern Hansen" <[EMAIL PROTECTED]> Cc: <[EMAIL PROTECTED]> Sent: Wednesday, February 05, 2003 9:57 PM Subject: Re: Large volume mailservers? > Well, outside of the caching nameserver, there are 5 other nameservers of > ours listed in resolv.conf to potentially use. > I'm using bind currently. If i lower the number of servers to 100 and remove > dnsbl and rhsbl, it still occurs. > > The other odd thing about it, is that i end up getting a reverse host: > > tcpserver: ok 4092 aaron.visual.com:209.123.16.34:25 > imo-r09.mx.aol.com:152.163.225.105::59644 > tcpserver: ok 4094 aaron.visual.com:209.123.16.34:25 > imo-r07.mx.aol.com:152.163.225.103::56597 > tcpserver: ok 4078 aaron.visual.com:209.123.16.34:25 > imo-d10.mx.aol.com:205.188.157.42::34229 > tcpserver: ok 4082 aaron.visual.com:209.123.16.34:25 > imo-d08.mx.aol.com:205.188.157.40::47055 > tcpserver: ok 4084 aaron.visual.com:209.123.16.34:25 > imo-d08.mx.aol.com:205.188.157.40::47076 > tcpserver: ok 4085 aaron.visual.com:209.123.16.34:25 > imo-d10.mx.aol.com:205.188.157.42::34276 > > but thats as far as it goes after a certain level.... > > root@aaron/# grep 4092 /var/log/qpsmtpd.log > tcpserver: pid 4092 from 152.163.225.105 > tcpserver: ok 4092 aaron.visual.com:209.123.16.34:25 > imo-r09.mx.aol.com:152.163.225.105::59644 > *sits* > > And i know that it does infact work, because messages are getting through > occasionally. > > tcpserver: pid 4004 from 64.12.138.2 > tcpserver: ok 4004 aaron.visual.com:209.123.16.34:25 > omr-m02.mx.aol.com:64.12.138.2::60508 > 4004 loading plugins from ./plugins > 4004 running plugin inbox_check_for_abuse > 4004 220 inbox.net ESMTP qpsmtpd 0.21-dev ready; send us your mail, but not > your spam. > 4004 dispatching EHLO omr-m02.mx.aol.com > 4004 250-inbox.net Hi omr-m02.mx.aol.com [64.12.138.2] > 4004 250-PIPELINING > 4004 250 8BITMIME > 4004 dispatching MAIL From:<> > 4004 full from_parameter: From:<> > 4004 from email address : [<>] > 4004 running plugin inbox_check_for_abuse > 4004 <> checking for abuse at ./plugins/inbox_check_for_abuse line 16, > <STDIN> line 2. > 4004 getting mail from <> > 4004 250 <>, sender OK - how exciting to get mail from you! > 4004 dispatching RCPT To:<[EMAIL PROTECTED]> > 4004 running plugin inbox_valid_user > 4004 550 invalid user found -> [EMAIL PROTECTED] > 4004 dispatching RSET > 4004 250 OK > tcpserver: end 4004 status 0 > > ----- Original Message ----- > From: "Ask Bjoern Hansen" <[EMAIL PROTECTED]> > To: "ross mueller" <[EMAIL PROTECTED]> > Cc: <[EMAIL PROTECTED]> > Sent: Wednesday, February 05, 2003 9:45 PM > Subject: Re: Large volume mailservers? > > > > On Wed, 5 Feb 2003, ross mueller wrote: > > > > > Things are great overall, pperl seems to lower the resident memory > > > usage dramatically, The biggest hangup i seem to have is the > > > initial dns lookup it seems. > > > > Does it happen even when there's just one connection? Try to see if > > the logs can pinpoint exactly which lookup is taking so long. > > > > If you run dnscache, then also try looking there for clues. Maybe > > something is misconfigured to do a lookup that's timing out. > > > > If you don't run dnscache, is the nameserver overloaded? > > > > > > - ask > > > > -- > > ask bjoern hansen, http://www.askbjoernhansen.com/ !try; do(); > > >
