Description

This article describes a minimal downtime upgrade procedure for an SRX345 chassis cluster.
The method upgrades one node at a time, allows traffic to move to the upgraded node, and then restores the cluster after both nodes have been updated.

Symptoms

This is an informational KB.

Use this procedure when:

  • An SRX345 cluster needs a software upgrade
  • The goal is to reduce service impact
  • A node-by-node upgrade is preferred

Solution

  • Perform the upgrade during a planned maintenance window using the following sequence:

    1. Disable the network interfaces on Node 1 and isolate it from the network.
    2. Break the control and fabric links to isolate the node from the cluster.
    3. Upgrade the software on Node 1 and reboot it.
    4. Validate that Node 1 is healthy and ready to forward traffic.
    5. Switch traffic from Node 0 to Node 1.
    6. Upgrade the software on Node 0 and reboot it.
    7. Reconfigure the control and fabric links on Node 0, then power it down.
    8. Reconfigure the control and fabric links on Node 1, then boot Node 0.
    9. Verify the cluster status after Node 0 rejoins.
    10. Re-enable the physical interfaces on Node 0.

    Summary

    A minimal downtime upgrade on an SRX345 cluster can be completed by isolating one node, upgrading it first, shifting traffic to it, upgrading the peer node, and then restoring full cluster operation.

 

Modification History

2024-09-21 : Article Created