I've tried this in various ways. First I used convert2rhel to convert the CentOS 7 VM to RHEL 7, then Leapp to move it to RHEL 8. Leapp did uninstall some of the Zimbra rpm's so I decided to 86 that method and switched to using new VM's. So I did exactly as stated below.
I now have a new RHEL 8 VM running Zimbra 10.0.13. That test vm seems to be working just fine except the Admin interface. Immediately after logging in I see a service (spell) is down, but zmcontrol status shows all services Running. The admin page is apparently old and marks all the service checks as stale. "data is stale" and according to the timestamp over 4 hours old.
And I cant refresh it. In other places in the Admin gui is get errors like:
Does anyone know what is going on here? this is the second incarnation of this test VM ending is this exact situation, just to be sure I did not mess up the directions given.
Thanks in advance.
I now have a new RHEL 8 VM running Zimbra 10.0.13. That test vm seems to be working just fine except the Admin interface. Immediately after logging in I see a service (spell) is down, but zmcontrol status shows all services Running. The admin page is apparently old and marks all the service checks as stale. "data is stale" and according to the timestamp over 4 hours old.
And I cant refresh it. In other places in the Admin gui is get errors like:
Code:
Server error encounteredMessage: system failure: Unable to read logger stats Error code: service.FAILURE Method: [unknown] Details:soap:ReceiverThanks in advance.
1) Zimbra 8.8.15 - CentOS 7
2) Zimbra 10.0.13 - CentOS 7
3) Zimbra 10.0.13 - CentOS 8
4) Zimbra 10.1.7 - CentOS 8
5) Zimbra 10.1.7 - RHEL 9
- Upgrade your current VM to 10.0.13 (RHEL7).
- Create a new VM with AlmaLinux/RockyLinux 8
- rsync the entire /opt/zimbra/ from old VM to new VM using these rsync options: -avxPAHX (important because Zimbra uses hard links for deduplication)
important: exlude the 'data/ldap/mdb/db/data.mdb' file because it is a sparse file (--exclude 'data/ldap/mdb/db/data.mdb')
- In the old VM, as zimbra user, take a copy of the sparse file like this:
Code:
$ mkdir /tmp/mdb$ mdb_copy data/ldap/mdb/db/data.mdb /tmp/mdb/
- Put /tmp/mdb/data.mdb from old VM to /opt/zimbra/data/ldap/mdb/db/ of new VM.
- Important: Take a backup of your /opt/zimbra/conf/localconfig.xml file. It will be overwritten later.
- Install 10.0.13 (RHEL8) on the new VM with the -s option (software only)
When it asks you if you want to empty /opt/zimbra before proceeding, say No. So you keep all your data.Code:
# ./install.sh -sRight now, you have a new VM with RHEL8 and 10.0.13.
- After it finishes, restore the localconfig.xml from the backup you took earlier and as root user fix all permissions:
Code:
# /opt/zimbra/libexec/zmfixperms -extended
You should verify that everything's working by starting the services
as zimbra user:If everything is normal, you can now proceed to upgrading to 10.1.7 (RHEL8) (without the -s this time)Code:
$ zmcontrol start
You must now repeat the whole process but this time you go from RHEL8 to RHEL9 and use the 10.1.7 (RHEL9) release instead.
Also, since at the last migration you used -s, zmsetup.pl will not be invoked, make sure to create the cronjob that will be missing from your system so that Zimbra stats work.Good luck.Code:
# cat /opt/zimbra/conf/crontabs/* > /var/spool/cron/zimbra# chmod 600 /var/spool/cron/zimbra
Statistics: Posted by JBravo — Fri Jun 27, 2025 10:18 pm