SLES 10 - New Install RC2 - Multiple issues.

[brucealeg]

1) During the install the first issue was the needing 0440 permissions on /etc/sudoers; which I fixed during the install.

If SLES maintains 640 (I think) on /etc/sudoers; why doesn't the SLES version of Zimbra handle this well?

2) During the end I received the following errors:

<< Setting services on zimbra...done.
Setting up zimbra crontab...cp: cannot stat `/var/spool/cron/tabs/zimbra': No such file or directory
done. >>

The file is there and populated, the contents are:

# DO NOT EDIT THIS FILE - edit the master and reinstall.
# (/tmp/crontab.zimbra installed on Mon Nov 26 15:54:15 2007)
# (Cron version V5.0 -- $Id: crontab.c,v 1.12 2004/01/23 18:56:42 vixie Exp $)
# ZIMBRASTART -- DO NOT EDIT ANYTHING BETWEEN THIS LINE AND ZIMBRAEND
#
# Log pruning
#
30 2 * * * find /opt/zimbra/log/ -type f -name \*.log\* -mtime +8 -exec rm {} \; > /dev/null 2>&1
#
# Status logging
#
*/2 * * * * /opt/zimbra/libexec/zmstatuslog
*/10 * * * * /opt/zimbra/libexec/zmdisklog
#
# Backups
#
# BACKUP BEGIN
# BACKUP END
#
# crontab.ldap
#
#
#
# crontab.store
#
# Log pruning
#
30 2 * * * find /opt/zimbra/mailboxd/logs/ -type f -name \*log\* -mtime +8 -exec rm {} \; > /dev/null 2>&1
0 2 * * * /opt/zimbra/libexec/zmroll_catalina > /dev/null 2>&1
#
# Purge messages
#
0 3 * * * /opt/zimbra/libexec/zmpurgemessages > /dev/null 2>&1
#
# Clean up the quarantine dir
#
0 1 * * * find /opt/zimbra/amavisd/quarantine -type f -mtime +7 -exec rm -f {} \; > /dev/null 2>&1
#
# Table maintenance
#
30 1 * * 7 /opt/zimbra/libexec/zmmaintaintables >> /dev/null 2>&1
#
# Report on any database inconsistencies
#
0 23 * * 7 /opt/zimbra/libexec/zmdbintegrityreport -m
#
# Monitor for multiple mysqld to prevent corruption


The other error is:

<< Setting up syslog.conf...cp: cannot stat `/etc/syslog.conf': No such file or directory
grep: /etc/syslog.conf: No such file or directory
syslogd: no process killed
Failed >>

Again, this file is there and populated:


local0.* -/var/log/zimbra.log
auth.* -/var/log/zimbra.log
mail.* -/var/log/zimbra.log


Are these bad errors or is something missing here?

3) The logger service starts and runs (zmcontrol status says it's up), but the admin page says it's off, so no data. I know this isn't a critical service, but this has been happening to me for awhile over different versions.

4) Zimbra runs fine for 15-30 minutes and then all the services just stop. This happened in RC1 and is still there in RC2. I'm not sure what log to read to determine why this is happening. Any advice is appreciated.


SLES 10 SP 1 - Zimbra OSE SLES 10 5.0 RC2 build

Thanks,
Bruce

[brucealeg]

From /var/log/mail.

This is where things seem to start going south.

Nov 26 16:57:58 zimbra postfix/postfix-script[32297]: refreshing the Postfix mail system
Nov 26 16:57:58 zimbra postfix/master[15642]: reload configuration /opt/zimbra/postfix-2.4.3.3/conf
Nov 26 16:58:07 zimbra zmmailboxdmgr[32497]: status requested
Nov 26 16:58:07 zimbra zmmailboxdmgr[32497]: status OK
Nov 26 16:59:43 zimbra zmmailboxdmgr[407]: threaddump requested
Nov 26 16:59:43 zimbra zmmailboxdmgr[21047]: sending SIQUIT to mailboxd/JVM process 21048
Nov 26 16:59:46 zimbra zmmailboxdmgr[414]: stop requested
Nov 26 16:59:46 zimbra zmmailboxdmgr[21047]: shutdown requested, sending TERM signal to 21048
Nov 26 16:59:46 zimbra zmmailboxdmgr[414]: waiting for manager process 21047 to die
Nov 26 16:59:48 zimbra zmmailboxdmgr[21047]: mailboxd/JVM process exited (waitpid expected 21048 got 21048)
Nov 26 16:59:48 zimbra zmmailboxdmgr[21047]: manager woke up from wait on mailboxd/JVM with pid 21048
Nov 26 16:59:49 zimbra zmmailboxdmgr[414]: manager process 21047 died, shutdown completed
Nov 26 16:59:58 zimbra postfix/postfix-script[530]: stopping the Postfix mail system
Nov 26 16:59:58 zimbra postfix/master[15642]: terminating on signal 15
Nov 26 16:59:59 zimbra zmmailboxdmgr[558]: threaddump requested
Nov 26 16:59:59 zimbra zmmailboxdmgr[558]: file /opt/zimbra/log/zmmailboxd.pid does not exist
Nov 26 16:59:59 zimbra zmmailboxdmgr[558]: assuming no other instance is running
Nov 26 16:59:59 zimbra zmmailboxdmgr[558]: no manager process is running
Nov 26 22:00:00 zimbra postfix/postqueue[561]: fatal: Queue report unavailable - mail system is down
Nov 26 17:00:02 zimbra zmmailboxdmgr[598]: stop requested
Nov 26 17:00:02 zimbra zmmailboxdmgr[598]: file /opt/zimbra/log/zmmailboxd.pid does not exist
Nov 26 17:00:02 zimbra zmmailboxdmgr[598]: assuming no other instance is running
Nov 26 17:00:02 zimbra zmmailboxdmgr[598]: no manager process is running
Nov 26 22:00:02 zimbra postfix/postqueue[623]: fatal: Queue report unavailable - mail system is down
Nov 26 17:00:09 zimbra zmmailboxdmgr[846]: status requested
Nov 26 17:00:09 zimbra zmmailboxdmgr[846]: file /opt/zimbra/log/zmmailboxd.pid does not exist
Nov 26 17:00:09 zimbra zmmailboxdmgr[846]: assuming no other instance is running
Nov 26 17:00:09 zimbra zmmailboxdmgr[846]: no manager process is running
Nov 26 22:00:30 zimbra postfix/postqueue[914]: fatal: Queue report unavailable - mail system is down
Nov 26 22:01:00 zimbra postfix/postqueue[923]: fatal: Queue report unavailable - mail system is down
Nov 26 22:01:30 zimbra postfix/postqueue[927]: fatal: Queue report unavailable - mail system is down
Nov 26 22:02:00 zimbra postfix/postqueue[931]: fatal: Queue report unavailable - mail system is down

[jholder]

Please attach your /opt/zimbra/log/zmmailboxd.out

Don't copy and paste, I'd like to see the whole log.

[brucealeg]

Here you go as request.

Thanks for your help,
Bruce

[jholder]

Can you turn off IMAP and POP3? and see if that helps? If not, please upload a new thread dump.

Looks like it's waiting on the IMAP SSL Server:

"ImapServer" prio=10 tid=0x082f5400 nid=0x3bf7 runnable [0xb4b79000..0xb4b7a150]
   java.lang.Thread.State: RUNNABLE
    at java.net.PlainSocketImpl.socketAccept(Native Method)
    at java.net.PlainSocketImpl.accept(PlainSocketImpl.java:384)
    - locked <0x7a590628> (a java.net.SocksSocketImpl)
    at java.net.ServerSocket.implAccept(ServerSocket.java:453)
    at java.net.ServerSocket.accept(ServerSocket.java:421)
    at com.zimbra.cs.tcpserver.TcpServer.run(TcpServer.java:175)
    at java.lang.Thread.run(Thread.java:619)

[brucealeg]

I sure can, we aren't going to use that anyhow. What's the best way to do that?

[jholder]

From the admin ui, you should be able to just uncheck them and restart zmmailboxd

su - zimbra
zmmailbodctl restart

I'm still fishing though the logs. There are a lot of locked threads, so it could be something else.

[brucealeg]

Sigh - I knew that. Just brain dead today. I just did it and will watch and see what happens. Takes 15-20 minutes or so to happen.

Thanks,
Bruce

[jholder]

Just looked through the log. Looks like multiple threads are locking, so it's very difficult to tell what's causing it. Can you send me you mailbox.log?

jholder@zimbra.com

Thanks

[brucealeg]

Here you go.