I would like to share an issue and the resolution. I dont think I've seen this before and Google dont give that much on it.
JUNOS 12.3X48-D85.1 will become active at next rebootWARNING: A reboot is required to load this software correctlyWARNING: Use the 'request system reboot' commandWARNING: when software installation is completeSaving package file in /var/sw/pkg/junos-12.3X48-D85.1.tgz ...Saving state for rollback ...error: [Oct 8 02:03:16]: Failed to install image on secondary node (error-code: 3.1)error: [Oct 8 02:03:16]: ISSU Aborted! Backup node maybe in inconsistent state, Please restore backup node[Oct 8 02:03:16]: ISSU aborted. But, both nodes are in ISSU window.Please do the following:1. Rollback the node with the newer image using rollback commandNote: use the 'node' option in the rollback commandotherwise, images on both nodes will be rolled back2. Make sure that both nodes (will) have the same image3. Ensure the node with older image is primary for all RGs4. Abort ISSU on both nodes5. Reboot the rolled back node
I've seen this once before, on this cluster during the same upgrade (from D60), that time we aborted and thought it was another reason for the issue.
Now when we had this again we tried executing ISSU with user root instead of admin account "myuser".
Then the ISSU worked perfect and was quick and pretty seamless, as expected.
Resolution as mentioned, run ISSU as the user "root" instead of custom admin account.