On Junos OS SRX platforms having chassis cluster configuration-synchronize configured, ISSU (In-Service Software Upgrade) gets aborted due to a configuration synchronization (config-sync) failure and the redundancy group (RG) priority is set to 0, preventing a successful failover during the ISSU process.
During ISSU, the secondary node's redundancy group priority is set to 0, and the config-sync monitor weight is set to 255. This causes a configuration synchronization failure, which in turn prevents failover and results in the ISSU process being aborted. This issue happens when "configuration-synchronize" is configured for chassis cluster.
On Node0
---------
<timestamp> node0: triggering issu failover for all the RGs to node1
<timestamp> rg 0 -remote- state: secondary priority: 0 <<<<< failover during ISSU failed because priority is 0
<timestamp> Abort ISSU!
<timestamp> Aborted ISSU due to error.
<timestamp> fab0 marked down
<timestamp> <hostname> mgd[63811]: UI_CLUSTER_ISSU_IN_PROGRESS: In progress; secondary node1 ready for failover.
<timestamp> <hostname> mgd[63811]: UI_CLUSTER_ISSU_ERROR_ABORT_STAY: Abort and stay in ISSU window; failover all RGs to node node1 failed (error-code: 7.1)
<timestamp> <hostname> mgd[63811]: UI_CLUSTER_ISSU_ERROR_ABORT_STAY: Abort and stay in ISSU window; but, both nodes are in ISSU window. Please follow the steps to recover the node
On Node1:
<timestamp> Current threshold for rg-0 is 0. Setting priority to 0. Failures: config-sync-monitoring
<timestamp> csmon failure, computed-weight 255, cs-mon-weight 255 <<<<<<<< config sync monitor weight is set to 255
<timestamp> Current threshold for rg-1 is -255. Setting priority to 0. Failures: cold-sync-monitoring, config-sync-monitoring
<timestamp> csmon failure, computed-weight 255, cs-mon-weight 255
<timestamp> <hostname> mgd[25704]: UI_CLUSTER_ISSU_IN_PROGRESS: In progress; secondary node0 ready for failover.
<timestamp> <hostname> mgd[25704]: UI_CLUSTER_ISSU_ERROR_ABORT_STAY: Abort and stay in ISSU window; failover all RGs to node node0 failed (error-code: 7.1)
<timestamp> <hostname> mgd[25704]: UI_CLUSTER_ISSU_ERROR_ABORT_STAY: Abort and stay in ISSU window; but, both nodes are in ISSU window. Please follow the steps to recover the node
Workaround
Disable config sync by setting the CLI command before ISSU and then delete this after ISSU completes.
set chassis cluster configuration-synchronize no-secondary-bootup-auto
Fix: Open Ticket to support team to get list of the fix images
Please consider for an upgrade to a fix releases