Jetty webapps not starting after server move

[Hivos]

Today I copied a complete backup of /opt/zimbra from a production server (ZCS 6.0.5 FOSS on Debian Etch 32bit) to a testserver, where I had first installed a dummy copy of ZCS 6.0.5 FOSS on Debian Etch 32bit. Somehow I can't get the webapps running. After a zmcontrol start they all (zimbra, zimbraAdmin, zimlet and service webapps) get stuck at the same point as shown in /opt/zimbra/log/zmmailboxd.out :

1834 WARN  [main] log - Failed startup of context org.mortbay.jetty.webapp.WebAppContext@1f4bcf7{/zimlet,/opt/zimbra/jetty-6.1.22.z6/webapps/zimlet}
java.io.FileNotFoundException: /opt/zimbra/jetty-6.1.22.z6/webapps/zimlet/WEB-INF/web.xml (Permission denied)

The web.xml file permissions are not OK apparently. Whenever I set these to 777 and then restart zmmailboxdctl the permissions are reset to 440 before the webapps start, etc etc.

All other services are running fine by the way. What I've already tried:

chown -R zimbra:zimbra /opt/zimbra
/opt/zimbra/libexec/zmfixperms

Which doesn't solve the problem. Anyone has a clue on where to start to get this loop sorted out?

[bdial]

try zmfixperms --extended

[Hivos]

try zmfixperms --extended

Thanks, but unfortunately it didn't resolve the issue.

I've just checked the production server, and the permissions there are exactly the same, so there must be something else preventing jetty from starting the webapps. Can't find much in the logfiles either. I purged all logfiles before starting zimbra, and these are the logfiles that show up after a zmcontrol start:

clamd.log
freshclam.log
httpd_access.log
httpd_error.log

[Fri Mar 05 02:57:03 2010] [notice] Apache/2.2.14 (Unix) PHP/5.2.12 configured -- resuming normal operations

myslow.log
mysql_error.log

100305 02:55:48  mysqld started
100305  2:55:50  InnoDB: Started; log sequence number 0 545686891
100305  2:55:50 [Note] /opt/zimbra/mysql/libexec/mysqld: ready for connections.
Version: '5.0.87-log'  socket: '/opt/zimbra/db/mysql.sock'  port: 7306  Source distribution

zmlogswatch.out
zmmtaconfig.log

FYI: the httpd_error.log and mysql_error.log contain only notices, no errors. The rest appears to contain nothing useful as well.

[legopost]

The problem is that the UID and GID of "zimbra" user on new server are not the same as on old server. Just fix it.

[conny]

The problem is that the UID and GID of "zimbra" user on new server are not the same as on old server. Just fix it.

I know this is an old thread, but I just wanted to say Thank You! I had the same problem and that hint solved it! Thanks!

[mitesh.choksi]

I found this issue affecting me too. I had installed Zimbra on a new system (uid=501,gid=501) and then restored backup of old server (uid=500/gid=500) using ZeXtras Backup module. Since my new install had uid 501 allocated and my backups expected 500. I changed the uid and gid in /etc/passwd and /etc/group to 500 and then also had to make a change in localconfig.xml in /opt/zimbra/conf directory to set the zimbra_uid and zimbra_gid to 500 too.

Hope this post helps others as it helped me.