Can't sent to non-local addresses

[smokes2345]

I finished installing scalix to CentOS4 yesterday and have been working on the configuration most of the day today. Mail can be sent to local users (from both local and non-local accounts) but none can be sent to non-local accounts. I've been searching the forums and internet most of the day for someone else who has experienced this problem and thought i had come across a topic that was helpful, however no solution was ever listed. I'm a bit of a linux noob but do know enough that when a mail is sent to a non-local account i am able to clear the Internet Mail Gateway (unix) queue. I need to get this server fully operational ASAP so any help is greatly appreciated and i will gladly post any logs that will be helpful although i may need a little help getting to them myself.

EDIT: A piece of information i just thought might be helpful, I tried to edit a file in ~scalix/sys and it wouldn't go, i opened up nautilus to see what the problem was and the sys folder wasn't even there. i'm not sure what kind of config files might be missing.

[mikethebike]

smoke

if the ~sys folder is missing, probably all the config files are missing.
Do you have that filesystem mounted?
Do you see anything in the logs (~logs), especially the fatal log

Mick

[smokes2345]

No, i'm not getting anywhere with ~logs. Do you mean ~scalix/logs? the only thing in the ~scalix folder after the install was a folder labeled "m2" whatever that is for.

EDIT: I have a /var/log folder, is that what i need to look for?

[Valerion]

Ah, look in /var/opt/scalix/??/s for the mailstore.

[smokes2345]

Ok, i have /var/opt/scalix/m2/s/msgpool/ , /var/opt/scalix/m2/s/msglist/, and /var/opt/scalix/m2/s/pmove.store/ but no mailstore. there are a few items in msgpool but they seem to be binary files as gedit complains when i try to open them. All the file names have 3 or 4 character strings for names. Those are my best geusses as to where the mailstore might be without something that says "mailstore".

[Valerion]

the "s" directory should have the following (maybe with one or two changes) under it. If it looks like this, then this is the mailstore itself. If there's lots of stuff missing (especially user and data) then your mailstore is definitely incomplete.

Code: Select all

drwxrwxr-x   5 scalix scalix 4.0K Jul 28  2005 access
drwxrwx--x   9 scalix scalix 4.0K Jan  2  2007 acl
drwxrwxr-x   4 scalix scalix 4.0K Jan 14 20:48 admin
drwxrwx---   3 scalix scalix 4.0K Jan  2  2007 bb
drwxrwx---   3 scalix scalix 4.0K Jan  2  2007 cda
drwxrwxr-x   2 scalix scalix 4.0K Nov 17 11:34 Copyright
drwxrwxr-x   2 scalix scalix 4.0K Nov 17 11:35 csetconv
drwxrwx--- 322 scalix scalix  12K Mar 19  2007 data
drwxrwx---   3 scalix scalix 4.0K Jan  2  2007 dblogs
drwxrwx--x   2 scalix scalix 4.0K Jan  2  2007 diag
drwxrwx---   3 scalix scalix 4.0K Jul 28  2005 dir
drwxrwx---   2 scalix scalix 4.0K Nov 17 11:36 dirchlog
drwxrwx---   2 scalix scalix 4.0K Jul 28  2005 dits
drwxrwx---   2 scalix scalix 4.0K Feb 16  2006 ds
drwxrwx---   2 scalix scalix  20K Jan 15 09:32 indexwork
drwxrwxr-x   3 scalix scalix 4.0K Jul 28  2005 is
drwxrwx--x   2 scalix scalix 4.0K Jan  2  2007 lic
drwxrwxr-x   2 scalix scalix 4.0K Nov 17 11:34 License
drwxrwxr-x   2 scalix scalix 4.0K Jan  2  2007 LicenseKeys
drwxrwxr-x   2 scalix scalix 4.0K Jan 15 00:01 logs
drwxrwx--x   2 scalix scalix 4.0K Jul 28  2005 msglists
drwxrwx--x   2 scalix scalix 4.0K Sep  8  2005 msgpool
drwxrwxr-x   2 scalix scalix 4.0K Jan  2  2007 nits
drwxrwxr-x  15 scalix scalix 4.0K Jan 19  2007 nls
drwxrwxr-x   3 scalix scalix  12K Jan 15 09:32 notifs
drwxrwxr-x   2 scalix scalix 4.0K Jan  2  2007 notify
drwxr-xr-x   2 root   root   4.0K Jan 14 21:35 oldDbLogs
drwxrwxr-x  51 scalix scalix 4.0K Nov 17 11:34 omhtml
drwxrwx--- 213 scalix scalix 4.0K Dec  6 11:32 orphans
drwxrwxr-x   2 scalix scalix 4.0K Aug 21 14:23 plugin
drwxrwx---   2 scalix scalix 4.0K Jan  2  2007 pmove.store
drwxrwx---   2 scalix scalix 4.0K Jan 14 20:48 pop3
drwxrwx--x   2 scalix scalix 4.0K Jul 28  2005 queues
drwxrwxr-x   2 scalix scalix 4.0K Nov 17 11:34 release.cst
drwxrwxr-x   8 scalix scalix 4.0K Nov 17 11:34 release.sys
drwxrwxr-x   2 scalix scalix 4.0K Nov 17 11:34 release.tou
drwxrwxr-x   2 scalix scalix 4.0K Jan  2  2007 req
drwxrwx---   2 scalix scalix 4.0K Aug 21 13:59 rules
drwxrwxrwx   2 root   root   4.0K Nov 17 11:34 rules.eg
drwxrwx--x   2 scalix scalix  20K Jan 14 21:37 scan.serv
drwxrwx--x   2 scalix scalix 4.0K Dec 30 02:15 scan.work
drwxrwxr-x   2 scalix scalix 4.0K Jan 14 20:48 se
drwxrwxr-x   2 scalix scalix 4.0K Jan  2  2007 structdb
drwxrwxr-x   2 scalix scalix 4.0K Jan 14 21:40 structlog
drwxrwxr-x  19 scalix scalix 4.0K Jan 14 20:48 sys
drwxrwx--x   4 scalix scalix 260K Jan 15 09:34 temp
drwxrwxrwx   2 scalix scalix 4.0K Jan 15 05:00 tmp
drwxrwxr-x   2 scalix scalix 4.0K Nov 17 11:35 toupper
drwxrwx---   3 scalix scalix 4.0K Dec 23  2005 trace
drwxrwx--x  34 scalix scalix 4.0K Oct 26 13:22 user
drwxrwxr-x   2 scalix scalix 4.0K Jan  2  2007 webusers
drwxrwx--x   2 scalix scalix 4.0K Jan  2  2007 xport.hold

[smokes2345]

from that list all i am missing is the access and orphans folder.

[Valerion]

Then your mailstore seems to be complete. Check the logs directory in here for the logs mentioned earlier, as well as sys. All Scalix configuration and data files live in this tree somewhere.

[smokes2345]

there are some errors in a file called fatal, most of them are probably related to my first attempts to clear a stuck message out of the queue. The last two may have something to do with the message although i doubt it as there have been more messages stuck than there are error messages. Check out the log at http://d.cl1p.net/scalix.fatal.log/

[ltward]

It looks like your queue manager is not running.

Do "omstat -a" and see if the Queue Manager is running.
If it is not, stop and restart Scalix (omshut, omrc) and see if teh Queue Manager starts up.

If Queue Manager fails to start when you cycle Scalix, pay special attention to any information displayed on your screen, and check your logs to see what errors show up there ("omshowlog -l 7 -p 5" is a good place to start).

[smokes2345]

only service that isn't running is something called "item structure server", queue manager says it is started and is a "non-stop" service.

[ltward]

Okay those are "old" entries so not relevant to your problem.

The very last entry, however, reports an om.dmon 1011.

Code: Select all

# omsolve -n om.dmon 1011
-------------------------------------------------------------------------------
Error Group: OM.DMON  Error Number: 1011

A queue name specified does not exist.

A request to update a queue was made but the queue could not be found.
This error may indicate an internal error within the queue manager.
Try restarting Scalix at the earliest opportunity.
-------------------------------------------------------------------------------


Have you restarted Scalix since those errors were logged? And the problem persists?

[ltward]

Oh, one more thing... I suspect the message is never getting put on your unix queue successfully, but a good thing to check is your sendmail logs. Does the message ever make it into sendmail? If not, that *really* points to the unix queue as the culprit!

What do you mean when you say in your original post

...i am able to clear the Internet Mail Gateway (unix) queue.

[smokes2345]

based on another post, i was able to write a simple shell script. It turns off the queue, kills any hung processes, then calls omomqdump to clear any stuck messages. After the user quits omqdump it then turns on the queue and runs omscan, commands are listed below:

omoff -d 0 -s unix
kill `pgrep unix.out`
(echos the days password for omqdump)
omqdump
omon -s unix
omscan -A -q -f

[smokes2345]

I suspected that sendmail was the problem, but i wasn't sure if there may have been some kind of configuration file i was missing since i had to create the "sys" directory. I've tried looking for sendmail logs but haven't been able to find them. i've confirmed that it is installed using yum. I've restarted the whole server several times since install, i'm fairly sure it's been restarted since that log entry.