smtp error

[philburress]

Anyone know what this error is:

xinetd[xxxxx]: service smtp, accept: Invalid argument (errno = 22)

I see a bunch of them in my logs. I've searched everywhere and can't find anything on it.

Thanks.

 

[ashleym]

I'm having the exact same problem.

My "/var/log/messages" log file is climbing in size by the minute - currently 1.72 GB - with the same error message you're receiving:

Mar 16 07:23:04 as xinetd[16448]: service ftp, accept: Invalid argument (errno = 22)
Mar 16 07:23:04 as xinetd[16448]: service poppassd, accept: Invalid argument (errno = 22)
Mar 16 07:23:04 as xinetd[16448]: service imap, accept: Invalid argument (errno = 22)

Also, I can't send or receive email - receiving "unable to connect to smtp server localhost:25"

I've heard that Watchdog may have been related to the problem so I uninstalled it, but that didn't fix it.

I've restarted all services and rebooted my server but I'm still stuck.

Please help..

 

[wagnerch]

Run these commands and dump the output here:

ps fuxwa |grep inet
cat /etc/xinetd.conf
ls -l /etc/xinetd.d/
grep "^service" /etc/xinetd.d/*


Also you may want to restart xinetd, to rule that out. I am wondering if you are a) running in a VPS, b) recently upgraded xinetd (patches applied?), c) if there is anything funky in /etc/xinetd.conf, or d) is xinetd simply running twice.

 

[ashleym]

a) VPS is installed
b) Last week my hoster mediatemple.net did apply some upgrades & patches
c) see below
d) no clue

---
Here's the out from the commands:
--

[root@as root]# ps fuxwa |grep inet
root 11507 0.0 0.0 1504 452 pts/0 S 20:04 0:00 \_ grep inet
root 17948 0.0 0.0 2004 892 ? S 19:31 0:00 xinetd -stayalive -pidfile /var/run/xinetd.pid


[root@as root]# cat /etc/xinetd.conf
#
# Simple configuration file for xinetd
#
# Some defaults, and include /etc/xinetd.d/

defaults
{
instances = 60
log_type = SYSLOG authpriv
log_on_success = HOST PID
log_on_failure = HOST
cps = 25 30
}

includedir /etc/xinetd.d


[root@as root]# ls -l /etc/xinetd.d/
total 25
-rw-r--r-- 1 root root 563 May 19 2004 chargen
-rw-r--r-- 1 root root 580 May 19 2004 chargen-udp
-rw-r--r-- 1 root root 364 Mar 16 19:19 courier-imapd
-rw-r--r-- 1 root root 294 Mar 16 19:19 courier-imapds
-rw-r--r-- 1 root root 364 Mar 16 19:19 courier-pop3d
-rw-r--r-- 1 root root 294 Mar 16 19:19 courier-pop3ds
-rw-r--r-- 1 root root 419 May 19 2004 daytime
-rw-r--r-- 1 root root 438 May 19 2004 daytime-udp
-rw-r--r-- 1 root root 341 May 19 2004 echo
-rw-r--r-- 1 root root 360 May 19 2004 echo-udp
-rw-r--r-- 1 root root 318 Aug 8 2005 finger
-rw-r--r-- 1 root root 286 Aug 24 2005 ftp_psa
-rw-r--r-- 1 root root 275 Aug 8 2005 ntalk
-rw-r--r-- 1 root root 355 Aug 30 2005 popa3d.rpmsave
-rw-r--r-- 1 root root 334 Aug 30 2005 popa3ds.rpmsave
-rw-r--r-- 1 root root 206 Mar 8 01:13 poppassd_psa
-rw-r--r-- 1 root root 317 Jan 14 2004 rsync
-rw-r--r-- 1 root root 314 May 19 2004 services
-rw-r--r-- 1 root root 392 Jan 31 2003 sgi_fam
-rw-r--r-- 1 root root 421 Mar 8 01:16 smtp.psa
-rw-r--r-- 1 root root 422 Mar 8 01:16 smtps.psa
-rw-r--r-- 1 root root 363 Jan 16 2004 swat
-rw-r--r-- 1 root root 263 Aug 8 2005 talk
-rw-r--r-- 1 root root 497 May 19 2004 time
-rw-r--r-- 1 root root 518 May 19 2004 time-udp


[root@as root]# grep "^service" /etc/xinetd.d/*
/etc/xinetd.d/chargen:service chargen
/etc/xinetd.d/chargen-udp:service chargen
/etc/xinetd.d/courier-imapd:service imap
/etc/xinetd.d/courier-imapds:service imaps
/etc/xinetd.d/courier-pop3d:service pop3
/etc/xinetd.d/courier-pop3ds:service pop3s
/etc/xinetd.d/daytime:service daytime
/etc/xinetd.d/daytime-udp:service daytime
/etc/xinetd.d/echo:service echo
/etc/xinetd.d/echo-udp:service echo
/etc/xinetd.d/finger:service finger
/etc/xinetd.d/ftp_psa:service ftp
/etc/xinetd.d/ntalk:service ntalk
/etc/xinetd.d/popa3d.rpmsave:service pop3
/etc/xinetd.d/popa3ds.rpmsave:service pop3s
/etc/xinetd.d/poppassd_psa:service poppassd
/etc/xinetd.d/rsync:service rsync
/etc/xinetd.d/services:service services
/etc/xinetd.d/sgi_fam:service sgi_fam
/etc/xinetd.d/smtp.psa:service smtp
/etc/xinetd.d/smtps.psa:service smtps
/etc/xinetd.d/swat:service swat
/etc/xinetd.d/talk:service talk
/etc/xinetd.d/time:service time
/etc/xinetd.d/time-udp:service time


I just stopped/started xinetd and I'm still having the same problem.

Also, when I run "/var/qmail/bin/qmail-qstat" it says the following:

messages in queue: 47
messages in queue but not yet preprocessed: 6

Thank you in advance for any help you can provide!! My hoster is saying that everything is fine..

 

[wagnerch]

Something is not quite right there... The pid of xinetd is 17948, but your errors were coming from another copy running as 16448. Try this:

ps w -p 16448
ls -l /proc/16448

Let's see what that process is. Did you reboot or anything like that?

 

[wagnerch]

Another thing that jumps out at me is courier imap/pop/et.al is running from xinetd. I am not sure which OS you are using, but mine runs as a backend daemon. What OS are you using?

Check if you have courier-imap in /etc/rc.d/init.d

and if you are using an RPM based OS then check this:

rpm -qa |grep courier

 

[ashleym]

Yes, I did reboot about an hour ago. I'm running RHEL 3.

I do have "/etc/rc.d/init.d/courier-imap" and it was updated when my host did the patches last week.

Here's the output:

[root@as root]# rpm -qa |grep courier
courier-imap-3.0.8-rhel3.build75050824.12
psa-courier-imap-add-7.5.4-rhel3.build75050902.17

The process 16448 currently isn't running as far as I can tell.

Another interesting thing is that the Plesk Admin panel won't let me stop or restart the "SMTP Server (QMail)" service.

 

[ashleym]

I don't want to speak to soon, but I think I fixed my problem.

From what I can tell, the latest Plesk upgrades that my host installed renamed the following files:

/etc/xinetd.d/smtp_psa
/etc/xinetd.d/smtps_psa

to:

/etc/xinetd.d/smtp.psa
/etc/xinetd.d/smtps.psa

By renaming them back and rebooting, I can now send/receive.

Only drawback is that I now have 43 emails in queue "/var/qmail/bin/qmail-qstat" that seem to be stuck.

Any ideas on how to get them?

 

[wagnerch]

I would also *move* the files courier* out of /etc/xinetd.d and then do a "service xinetd reload". Verify that courier-imap is running by doing "service courier-imap status", it should say it is running. If it isn't then do "service courier-imap start" and "chkconfig courier-imap on".

 

[ashleym]

1. I moved the files out of the directory
2. ran "service xinetd reload" successfully
3. ran "service courier-imap status" and it wasn't running
4. ran "service courier-imap start" and it wouldn't start
5. ran "chkconfig courier-imap on" gave "ok"

Any thoughts on how to get my queued emailed?

"/var/qmail/bin/qmail-qstat" shows 52 messages in queue.

Thanks again for your help!

 

[wagnerch]

Hmm, sounds like your ISP humped up a few things -- sounds like imap/pop3/etc may be screwed up?

You may need to reinstall the qmail package, do a forced installation.

service qmail stop
rpm -ivh /var/psa/PSA_7.5.4/rpm_FedoraCore_2/base/psa-qmail-1.03-fc2.build75050824.12.i586.rpm --force
service qmail start

And see what happens. It could always be worse .

 

[ashleym]

Is that the correct rpm for RHEL3?

By doing this, is there a chance that things could get worse?

 

[wagnerch]

You shouldn't need to do the drweb-qmail part, just psa-qmail. My path was an example, you may have to look around for it or download it from SWsoft. In any event it is documented below:

http://faq.sw-soft.com/article_22_252_en.html


Normally when qmail just stops functioning it is related to the queue structures. There is a queue-fix tool, but the above is the easiest solution.