Source release only has swap space of 2 GB, CM 15 upgrade

iverson.justin · ‎06-20-2025

Has anyone ran into the precop check for Source release only has swap space of 2 GB, which is not enough for destination release 15. We are running running 110 Disk CM 12.5 and our precheck is getting this. Should we try to upgrade and see what happens? I didn't know the system look into swap partition and there were limits.

Michael.Heimann · ‎06-20-2025

Hi, this is an unsupported update track. You should probably backup, redeploy with latest 12.5 OVAs, install and restore from backup before doing the upgrade. This will give you 4 GB Swap.

This being said, you can follow an unsupported path I posted here:

https://community.cisco.com/t5/unified-communications-infrastructure/cucm-11-5-1-upgrade-to-14-1su3-preupgrade-failed-check-1-6/m-p/5122585#M177529

Basically, it's increasing the swap partition with a live CD and changing ext3 to ext4 in fstab. If you understand that, it's not a big deal.

iverson.justin · ‎06-20-2025

Thanks good info. Do you know if we did the Data Export and Import Feature can we do node by node or is the cluster down and no changes can be made until all versions are on 15? Seems like a big outage window if we can't do a node at a time. Thanks

Nithin Eluvathingal · ‎06-20-2025

When performing a fresh install with the data export option, the process is done node by node. First, the Publisher is installed, followed by the Subscriber. During each node’s installation, the exported data is imported as well.

If your question is whether you can have some nodes on version 12.5 and others on version 15, and make changes in this setup, it might be possible, but it’s not recommended. For a fresh install, it’s best to complete the entire cluster, then migrate phones and services to the new cluster.

Regarding maintenance window concerns, you won’t experience a noticeable outage. Phones will reregister to nodes based on the CMgroup configurations.

Alternatively, set up a lab environment to build the new cluster. During the maintenance window, disconnect the old cluster from the network and connect the new one to minimize downtime.