Exim not starting :(

F

francis_joe

Verified User
Joined
Aug 1, 2004
Messages
10
Location
India
We are unable to start exim on our server.

Below are the details which i got from my investigation.

# /etc/init.d/exim status
exim dead but pid file exists


# exim -bd -d+all
00:17:44 3828 Exim version 4.34 uid=0 gid=0 pid=3828 D=ffffffff
Berkeley DB: Sleepycat Software: Berkeley DB 3.3.11: (July 12, 2001)
Support for: iconv() Perl OpenSSL
Lookups: lsearch wildlsearch nwildlsearch iplsearch cdb dbm dbmnz
Authenticators: cram_md5 plaintext spa
Routers: accept dnslookup ipliteral manualroute queryprogram redirect
Transports: appendfile/maildir/mailstore/mbx autoreply lmtp pipe smtp
00:17:44 3828 changed uid/gid: forcing real = effective
00:17:44 3828 uid=0 gid=0 pid=3828
00:17:44 3828 auxiliary group list: <none>
00:17:44 3828 --Malloc 0x8345d40 8200 store.c 139 8200 4265
00:17:44 3828 ---1 Get 0x8345d48 16 string.c 345
00:17:44 3828 ---1 Get 0x8345d58 24 string.c 345
00:17:44 3828 ---1 Get 0x8345d70 24 string.c 345
00:17:44 3828 ---1 Get 0x8345d88 8 string.c 345


....
..

00:17:44 3828 ---0 Get 0x834d038 8 string.c 345
00:17:44 3828 originator: uid=0 gid=0 login=root name=root
00:17:44 3828 ---0 Get 0x834d040 8 daemon.c 932
00:17:44 3828 ---0 Get 0x834d048 64 host.c 451
00:17:44 3828 ---0 Get 0x834d088 8 daemon.c 1053
00:17:44 3828 ---0 Get 0x834d090 800 daemon.c 1072
00:17:44 3828 listening on all interfaces (IPv4) port 25
00:17:44 3828 pid written to /var/run/exim.pid
00:17:44 3828 changed uid/gid: running as a daemon
00:17:44 3828 uid=8 gid=12 pid=3828
00:17:44 3828 auxiliary group list: 12
00:17:44 3828 ---1 Get 0x834c978 24 string.c 345
00:17:44 3828 LOG: MAIN
00:17:44 3828 exim 4.34 daemon started: pid=3828, no queue runs, listening for SMTP on port 25 (IPv4)
00:17:44 3828 LOG: PANIC DIE
00:17:44 3828 Cannot open main log file "/var/log/exim/mainlog": Permission denied: euid=8 egid=12
00:17:44 3828 2005-02-04 00:17:44 exim 4.34 daemon started: pid=3828, no queue runs, listening for SMTP on port 25 (IPv4)
00:17:44 3828 2005-02-04 00:17:44 Cannot open main log file "/var/log/exim/mainlog": Permission denied: euid=8 egid=12
00:17:44 3828 exim: could not open panic log - aborting: see message(s) above
00:17:44 3828 search_tidyup called
00:17:44 3828 >>>>>>>>>>>>>>>> Exim pid=3828 terminating with rc=1 >>>>>>>>>>>>>>>>

# ll -d /var/log/exim
drwxr-xr-x 2 mail mail 4096 Jun 13 2004 /var/log/exim

# ll /var/log/exim
total 0
-rw-r--r-- 1 mail mail 0 Feb 3 23:54 mainlog
-rw-r--r-- 1 mail mail 0 Feb 3 23:54 paniclog
-rw-r--r-- 1 mail mail 0 Feb 3 23:54 processlog
-rw-r--r-- 1 mail mail 0 Feb 3 23:54 rejectlog


Could somebody please guide me as soon as possible.
 
From my Unix/Linux knowledge, I would say either Exim is running as the wrong user, or more likely, you need to chmod your log files. Try:

chmod -R 666 /var/log/exim

or

chmod -R 777 /var/log/exim

Louie
 
Hmm....your exim log files have the same permissions as mine and exim works ok for me.

Maybe exim isn't running as the user 'mail'???

Louie
 
Hi Francis,

Please run the following commands on your server and post the results back here:


grep ^mail /etc/passwd | cut -f1,3-4 -d:
ls -lad /var/log
ls -lad /var/log/exim
ls -l /var/log/exim

With the results from those commands, we might be able to figure out why exim can open the log files. Also, please check that your /var/log directory is not located on a read-only mount for some reason. If you are the root user, does the following command work?

touch /var/log/exim/testlog

Regards,
Greg
 
server# ls -lad /var/log
drwxr-xr-x 6 root wheel 1536 Mar 12 14:00 /var/log
server# ls -lad /var/log/exim
drwxr-xr-x 2 mail mail 512 Mar 24 15:00 /var/log/exim
server# ls -l /var/log/exim
total 42010
-rw------- 1 mail mail 35749841 Mar 24 15:00 mainlog
-rw------- 1 mail mail 5529230 Mar 12 00:00 mainlog.0
-rw------- 1 mail mail 1257870 Mar 11 00:00 mainlog.1
-rw------- 1 mail mail 11147 Mar 10 08:00 mainlog.2
-rw------- 1 mail mail 10240 Mar 24 15:00 paniclog
-rw------- 1 mail mail 746 Mar 12 00:00 paniclog.0
-rw------- 1 mail mail 1161 Mar 11 00:00 paniclog.1
-rw------- 1 mail mail 1100 Mar 10 08:00 paniclog.2
-rw------- 1 mail mail 311259 Mar 24 14:09 rejectlog
-rw------- 1 mail mail 10208 Mar 12 00:00 rejectlog.0
-rw------- 1 mail mail 16399 Mar 11 00:00 rejectlog.1
-rw------- 1 mail mail 546 Mar 10 08:00 rejectlog.2
-rw-r--r-- 1 root mail 0 Mar 24 15:00 testlog
 
We're most likey running different distributions, because CentOS doesn't by default have a wheel group.

But on our CentOS servers, our files inside the /var/log/exim directory are all chmod 644 owned by mail:mail. Your are all owned by mail:mail, but are chmod 600.

Jeff
 
I changed their chmod settings but my problem still continues.

How can I clean my var directory ?
 
I have no idea what you mean by clean my var directory.

Start off by explaining which OS distribution you're using and perhaps someone who runs that distribution can help you.

Jeff
 
Because the error caused by the full of drive.

I cleaned my var log directory and evertying is ok for now.
 
You can always delete log files followed by .0 where that 0 can represent any number.

You should always start with the highest numbers first; for example delete *.4 before *.3, etc.

Jeff
 
You must log in or register to reply here.
Back
Top