failed upgrade from 5.0.11 to 6.0.6

[admin1]

All seemed fine until :

Restoring existing configuration file from /opt/zimbra/.saveconfig/localconfig.xml...done
Operations logged to /tmp/zmsetup.04192010-004026.log
Upgrading from 5.0.11_GA_2695 to 6.0.6_GA_2324
Stopping zimbra services...done.
Verifying /opt/zimbra/conf/my.cnf
ERROR 1033 (HY000) at line 1: Incorrect information in file: './zimbra/config.frm'
Mon Apr 19 00:41:00 2010: Error while running '/opt/zimbra/bin/mysql --user=zimbra --password=xnWjGf0pHOUGmMvOEWEH_e0B8e9wM8 --database=zimbra --batch --skip-column-names'.

What might cause this?

Thanks

[azeem]

I'm haveing the same problem when upgrading from 5.0.14 to 6.0.7.1

I have seen a few references to this error, but no real solution. Any help is greatly appreciated


Arnljot

[iway]

Have you opened a support case?

[azeem]

I'm running the Open Source version, so I really need community help. All my other upgrades over the years have run without fault. Luckily I have a snapshot of the VM running the Zimbra server, but I hope I'll get the upgrade running before monday.

[phoenix]

Originally Posted by azeem

I'm running the Open Source version, so I really need community help. All my other upgrades over the years have run without fault. Luckily I have a snapshot of the VM running the Zimbra server, but I hope I'll get the upgrade running before monday.

Try this first: [SOLVED] Upgrade from 6.0.1 to 6.0.2 fails

When you've finished the upgrade please update your forum profile (it's somewhat out of date ) with the output of the following:

zmcontrol -v

[azeem]

Not so easy unfortunately. The md5sum checks out.

It seems the SQL server is already running. I tried to run the ./install.sh one more time, and this time I get the following error when verifying db integrity:

Do you want to verify message store database integrity? [Y] y
Verifying integrity of message store databases. This may take a while.
Starting mysqld...done.
/opt/zimbra/mysql/bin/mysqlcheck: Got error: 2002: Can't connect to local MySQL server through socket '/opt/zimbra/db/mysql.sock' (2) when trying to connect
No errors found
command failed

This did not happen during the first install/upgrade.


Also, when I start the sql server manually, I get the same error message as I get during install :

mail2:~# /opt/zimbra/bin/mysql --user=zimbra --password=<my password> --database=zimbra --batch --skip-column-names
ERROR 2002 (HY000): Can't connect to local MySQL server through socket '/opt/zimbra/db/mysql.sock' (2)

However, doing a '/opt/zimbra/bin/mysql.server status' tells me the mysq server is not running. Trying to start it with '/opt/zimbra/bin/mysql.server start' works without error, but status is stil 'not running' and the following is logged in /opt/zimbra/log/mysql_error.log

100620 12:35:48 mysqld started
100620 12:35:50 InnoDB: Started; log sequence number 0 565847811
100620 12:35:50 [ERROR] Can't start server: Bind on TCP/IP port: Address already in use
100620 12:35:50 [ERROR] Do you already have another mysqld server running on port: 7306 ?
100620 12:35:50 [ERROR] Aborting

100620 12:35:50 InnoDB: Starting shutdown...
100620 12:35:51 InnoDB: Shutdown completed; log sequence number 0 565847811
100620 12:35:51 [Note] /opt/zimbra/mysql/libexec/mysqld: Shutdown complete

100620 12:35:51 mysqld ended

When I do a 'ps aux | grep sql' , I get:

zimbra 11793 0.0 0.0 3164 1644 ? S 01:19 0:00 /bin/sh /opt/zimbra/mysql/bin/mysqld_safe --defaults-file=/opt/zimbra/conf/my.cnf --external-locking --log-error=/opt/zimbra/log/mysql_error.log --ledir=/opt/zimbra/mysql/libexec
zimbra 11839 0.0 2.2 569064 46796 ? Sl 01:19 0:00 /opt/zimbra/mysql/libexec/mysqld --defaults-file=/opt/zimbra/conf/my.cnf --basedir=/opt/zimbra/mysql --datadir=/opt/zimbra/db/data --pid-file=/opt/zimbra/db/mysql.pid --skip-external-locking --port=7306 --socket=/opt/zimbra/db/mysql.sock --external-locking
root 9838 0.0 0.0 2856 688 pts/0 R+ 12:19 0:00 grep sql

This is obviously a different binary.....

As you can tell, I have no clue about mysql, and some of the suff I'm trying is probably way off. However, I believe something is not right on the database side of things.

I have not had any db problems in the past. I run Zimbra on a VM on Debian 4 OS. The VM has 2 GB of RAM.

[iway]

You could try to kill all mysql processes befor doing the upgrade. kill process-id

[azeem]

Here is the startup sequence with errors:

Starting ldap...Done.
Unable to determine enabled services from ldap.
Enabled services read from cache. Service list may be inaccurate.
Starting logger...Failed.
Starting logswatch...ERROR: service.FAILURE (system failure: unable to lookup server by name: mail2.platinium.no message: [LDAP: error code 49 - Invalid Credentials]) (cause: javax.naming.AuthenticationException [LDAP: error code 49 - Invalid Credentials])
zimbra logger service is not enabled! failed.


Starting mailbox...Done.
Starting antispam...Done.
Starting snmp...Done.
Starting spell...Done.
Starting mta...Failed.
Starting zmmtaconfig...zmmtaconfig is already running.
Starting saslauthd...saslauthd[8301] :set_auth_mech : failed to initialize mechanism zimbra
failed.
zmsaslauthdctl failed to start


Starting stats...Done.

[phoenix]

Originally Posted by azeem

I have not had any db problems in the past. I run Zimbra on a VM on Debian 4 OS. The VM has 2 GB of RAM.

What sort of a VM is this, xen, vserver, VMware etc.?

Shutdown Zimbra and check if there's any of the services still running and kill them, when you've got rid of them all try the upgrade again. I'd suggest you download the current 6.0.7.1 for the upgrade.

Before you run the upgrade let's also see the output of the following commands (run on the Zimbra server):

cat /etc/hosts
cat /etc/resolv.conf
dig yourdomain.com mx
dig yourdomain.com any
host `hostname` <- use that exact command with backticks not single quotes

[azeem]

Thanks for your input Phoenix.

I'm running on Xen, and have been for several years. I'm using FOSS 6.0.7.1 to upgrade. AFter stopping Zimbra, two mysql proccesses are still running. I killed them both, and did the upgrade again, with the same result.

Here are my dns details:

mail2:~# cat /etc/hosts
127.0.0.1 localhost.platinium.no localhost
192.168.0.9 mail2.platinium.no mail2

mail2:~# cat /etc/resolv.conf
search tveterveien.local
nameserver 192.168.0.10

mail2:~# dig platinium.no mx

; <<>> DiG 9.3.4-P1.2 <<>> platinium.no mx
;; global options: printcmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 30509
;; flags: qr aa rd ra; QUERY: 1, ANSWER: 0, AUTHORITY: 1, ADDITIONAL: 0

;; QUESTION SECTION:
;platinium.no. IN MX

;; AUTHORITY SECTION:
platinium.no. 3600 IN SOA winsrv01.tveterveien.local. hostmaster.tveterveien.local. 11 900 600 86400 3600

;; Query time: 4 msec
;; SERVER: 192.168.0.10#53(192.168.0.10)
;; WHEN: Sun Jun 20 17:01:01 2010
;; MSG SIZE rcvd: 103

mail2:~# dig platinium.no any

; <<>> DiG 9.3.4-P1.2 <<>> platinium.no any
;; global options: printcmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 9277
;; flags: qr aa rd ra; QUERY: 1, ANSWER: 2, AUTHORITY: 0, ADDITIONAL: 1

;; QUESTION SECTION:
;platinium.no. IN ANY

;; ANSWER SECTION:
platinium.no. 3600 IN NS winsrv01.tveterveien.local.
platinium.no. 3600 IN SOA winsrv01.tveterveien.local. hostmaster.tveterveien.local. 11 900 600 86400 3600

;; ADDITIONAL SECTION:
winsrv01.tveterveien.local. 3600 IN A 192.168.0.10

;; Query time: 0 msec
;; SERVER: 192.168.0.10#53(192.168.0.10)
;; WHEN: Sun Jun 20 17:01:09 2010
;; MSG SIZE rcvd: 133

mail2:~# host mail2.platinium.no
mail2.platinium.no has address 192.168.0.9
mail2.platinium.no mail is handled by 10 mail2.platinium.no.
mail2:~#

The dig output looks incomplete, but the server is running in a local domain while hosting mail for two public domains. This has worked all the way up to my upgrade attempt.

Here you see the public mx record which points to our external virus cleaner:

Your Input platinium.no
Results
(MX) Pref 10, dkcphmx70.softcom.dk.
(MX) Pref 10, dkcphmx68.softcom.dk.
(MX) Pref 10, deframx24.softcom.dk.