06-26-2024 12:29 PM
Currently I have a 2 node CUCM 12.5 cluster running on ESXi 6.5 that I'm looking to get to version 15. Based on the v15 upgrade guide this is possible, but CUCM v15 isn't supported on ESXi 6.5, or at least not tested. Is the best option to try to do a fresh install of v15 on a 8.0 ESXi cluster using the data-import utility? Or would it be better to get ESXi to 6.7, then upgrade the cluster to 14, migrate to the 8.0 ESXi environment and then upgrade again to v15?
Would these be the only supported ways to upgrade (other than PCD)?
02-03-2025 01:02 AM - edited 02-03-2025 01:02 AM
True, but it depends what you are doing. As @Roger Kallberg mentions, it is very helpful migrating a selected set of users. Another thing is that the voicemail database changes more then the configuration database does in CUCM. What I would do when migrating the entire system was to build the new system on the target version with the user configs, but not bring over the voicemail messages. That would mean a system ready to receive messages at cutover time, and I could just change the destination for the appropriate route patterns to the SIP integration. Then I could run another COBRAS export after the existing system was no longer active including the messages, and import that over top of the new system. That would get messages and last minute config changes like password and greetings the users had done at the last minute. I did always try to NOT bring over the messages since saved ones reappear as new when you migrate them which can confuse the user. If you leave the old system accessible for a while via a different destination #, the users could get to old messages for a while.
Everything you have said about the newer methods is totally valid @Nithin Eluvathingal , but I have used the methods I described for years so I am still comfortable doing it that way. Old dogs and new tricks, I guess!
11-26-2024 07:26 AM - edited 11-26-2024 07:37 AM
Clear
02-04-2025 02:16 AM - edited 02-04-2025 02:27 AM
I'm sure the most correct solution would be, for the beginning, to run ciscocm.preUpgradeCheck-00043.cop.sha512 for CUCM & IMP and for CUC ciscocm.cuc_preUpgradeCheck-007.cop.sha512
02-04-2025 03:26 AM
That would surely show what needs to be taken into account for the upgrade to succeed , but it wouldn't itself solve anything for you. That's on the person reading the output from the pre upgrade checks.
02-04-2025 04:06 AM - edited 02-04-2025 04:07 AM
In short, my path would be like this:
1 Upgrading VMware to 8U1
2 Changing the VM's compatibility to 21
3 Changing the VM's OS version to Other Linux 4 -x64 (for 15, CentOS 7 for 14)
4 Check run preUpgrade cop file and if all PASS
5 utils system upgrade cluster initiate
02-04-2025 04:40 AM
FYI #3 in your list should be done after the upgrade has been done and the switch to new version.
This outtake is from this presentation by Justin Jordan at Cisco Live. For more details please see this link. Upgrade to Cisco Unified Communications Manager Release15 with Confidence!
02-04-2025 04:56 AM - edited 02-04-2025 04:58 AM
For All )
For example, run preUpgrade.cop on VMware 6.7U1 with 14.0.1.12900-161 (SU2)
1.9 FAIL OS and Infrastructure Checks
FAIL: Source release is on ESXi version 6, lower than 7.0 which is not
supported in release 15.
Upgrade to a supported ESXi release (see virtualization docs).
P.S. I mean, everyone really needs to read and then do it, as you wrote above
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide