Periodically my ispconfig3 server stops receiving and sending mail. The mail queue shows: 43E148C8001 6804 Thu Jan 14 09:53:18 [email protected] (connect to 127.0.0.1[127.0.0.1]:10024: Connection refused) [email protected] This happens for every mail attempting to send. Restarting postfix doesn't help, but restarting the server does. The server seems to be attempting to send alerts to [email protected]: but unfortunately its set to send to the wrong email address. Where do I go to change this? Thanks
Here's some more log info that may or may not be relevant: From mail err-log: Jan 13 21:28:32 ispc01 amavis[30094]: (30094-20) (!!)TROUBLE in check_mail: parts_decode_ext FAILED: file(1) utility (/usr/bin/file) error: run_command (open pipe): Can't fork at /usr/lib/perl/5.10/IO/File.pm line 66, line 121. at /usr/sbin/amavisd-new line 2892, line 121. From mail warn-log: Jan 13 21:28:32 ispc amavis[30094]: (30094-20) (!!)TROUBLE in check_mail: parts_decode_ext FAILED: file(1) utility (/usr/bin/file) error: run_command (open pipe): Can't fork at /usr/lib/perl/5.10/IO/File.pm line 66, line 121. at /usr/sbin/amavisd-new line 2892, line 121. Jan 13 21:28:32 ispc amavis[30094]: (30094-20) (!)PRESERVING EVIDENCE in /var/lib/amavis/tmp/amavis-20100113T212832-30094 Jan 13 21:28:33 ispc amavis[2431]: (!)Net::Server: 2010/01/13-21:28:33 Bad fork [Cannot allocate memory]\n at line 166 in file /usr/share/perl5/Net/Server/PreForkSimple.pm
So this seems to be a memory issue. Specifically with amavis. I've currently got 1GB of RAM. Seems like that should be enough to run. I don't have that much mail going on here. How about disabling amavis? I use a hardware device for mail scanning anyway.
It seems no matter how much memory I throw at this thing it invariably chews it up. free total used free shared buffers cached Mem: 2076452 1809352 267100 0 12312 140512 -/+ buffers/cache: 1656528 419924 Swap: 0 0 0 I upped the RAM from 1GB to 2GB and its now approaching capacity once again.
Just to recap, I'm asking two things here: 1. how to change the root mail address alerts are sent to 2. is it possible to completely disable any perl script antivirus mail scanning? I believe that's whats failing and creating a disconnect. Thanks again.
I've worked through permenently disabling amavis and clamav. Things look better. I'll post back if this doesn't correct the issue.
Your server had not enough ram to run a spam and antivirus filter for the emails, so it might be indeed better to deactivate them.
Yes, the memory load is much more consistent and minimal now after disables those services. How about change the address though? I'm still getting mail queued up trying to deliver to the root user of the ispconfig server address that doesn't exist: Code: 1B779622C7 11666 Fri Jan 15 06:51:25 MAILER-DAEMON (connect to does.not-exist.com[45.45.45.45]:25: Connection timed out) [email][email protected][/email] 2DC856202C 9589 Mon Jan 11 06:28:42 [email][email protected][/email] (delivery temporarily suspended: connect to does.not-exist.com[45.45.45.45]:25: Connection timed out) [email][email protected][/email] I'm guessing this must be alerts of some sort and I should configure an address that actually exists!
Code: /home/hidden# netstat -tap Active Internet connections (servers and established) Proto Recv-Q Send-Q Local Address Foreign Address State PID/Program name tcp 0 0 *:mysql *:* LISTEN 2730/mysqld tcp 0 0 *:submission *:* LISTEN 2904/master tcp 0 0 localhost.localdo:spamd *:* LISTEN 2382/spamd.pid tcp 0 0 *:sunrpc *:* LISTEN 1935/portmap tcp 0 0 *:http-alt *:* LISTEN 2018/apache2 tcp 0 0 *:www *:* LISTEN 2018/apache2 tcp 0 0 *:36146 *:* LISTEN 1946/rpc.statd tcp 0 0 *:ftp *:* LISTEN 27482/pure-ftpd (SE tcp 0 0 hidden.hidden:domain *:* LISTEN 2821/mydns tcp 0 0 localhost.locald:domain *:* LISTEN 2821/mydns tcp 0 0 *:ssh *:* LISTEN 2388/sshd tcp 0 0 *:smtp *:* LISTEN 2904/master tcp 0 0 *:https *:* LISTEN 2018/apache2 tcp 0 0 localhost.localdoma:637 *:* LISTEN 2157/famd tcp 0 0 localhost.localdo:mysql localhost.localdo:48322 ESTABLISHED 2730/mysqld tcp 0 0 hidden.hidden:smtp hidden.hidden:41836 TIME_WAIT - tcp 0 152 hidden.hidden:ssh hidden.hidden.co:57147 ESTABLISHED 9118/sshd: matador tcp 0 0 localhost.localdo:46333 localhost.localdo:mysql TIME_WAIT - tcp 0 0 hidden.hidden:smtp hidden.hidden:41883 TIME_WAIT - tcp 0 0 localhost.localdo:mysql localhost.localdo:46334 ESTABLISHED 2730/mysqld tcp 0 0 localhost.localdo:mysql localhost.localdo:48319 ESTABLISHED 2730/mysqld tcp 0 0 localhost.localdo:48319 localhost.localdo:mysql ESTABLISHED 8633/smtpd tcp 0 0 localhost.localdo:46334 localhost.localdo:mysql ESTABLISHED 9142/proxymap tcp 0 0 localhost.localdo:48320 localhost.localdo:mysql ESTABLISHED 8633/smtpd tcp 0 0 localhost.localdo:mysql localhost.localdo:48320 ESTABLISHED 2730/mysqld tcp 0 0 localhost.localdo:48322 localhost.localdo:mysql ESTABLISHED 8633/smtpd tcp6 0 0 [::]:imaps [::]:* LISTEN 2274/couriertcpd tcp6 0 0 [::]:pop3s [::]:* LISTEN 2299/couriertcpd tcp6 0 0 [::]:pop3 [::]:* LISTEN 2285/couriertcpd tcp6 0 0 [::]:imap2 [::]:* LISTEN 2258/couriertcpd tcp6 0 0 [::]:ftp [::]:* LISTEN 27482/pure-ftpd (SE tcp6 0 0 localhost:domain [::]:* LISTEN 2821/mydns tcp6 0 0 [::]:ssh [::]:* LISTEN 2388/sshd
That looks ok - Postfix is listening on all interfaces. So I think your firewall might block port 25.
Maybe I was unclear. The problem is that ISPconfig is trying to send to the wrong address and I don't know where to change the root address that alerts are sent to. The current address does not exist. I must have entered it during installation incorrectly.
You can set up an alias for root in /etc/aliases. Run Code: newaliases afterwards and restart Postfix.
