GFI Support Home Start a conversation Sign in
Start a conversation
  1. Exinda
  2. Articles
  3. Public Articles

Troubleshooting "Split Brain" Issues in Exinda Clusters After Upgrade

Overview

After upgrading Exinda devices from version 7.5.4 to 7.5.6, you may encounter a "split brain" scenario where both devices in a cluster show as master, leading to unstable connections and frequent logouts from the cluster IP. This article outlines steps to troubleshoot and resolve these issues.

Information

To resolve the "split brain" issue in your Exinda cluster after upgrading to version 7.5.6, follow these steps:

  1. Verify Cluster Configuration:
    • Use the following CLI commands to check the cluster configuration: 
      show cluster global brief
show cluster sync {acceleration|optimizer|monitor|compression}
show appliances
  2. Reboot and Re-establish Cluster:
    • Reboot both devices and wait for the cluster to re-establish.
    • If the issue persists, disable and re-enable the cluster using these commands: 
      shell
en
config t
no cluster enable
  3. Consider Downgrading to Version 7.5.5:
    • If problems continue, downgrade to version 7.5.5 by switching boot partitions and rebooting the appliance.
  4. Modify Auto Logout Timeout:
    • Increase the auto logout timeout settings by navigating to Configuration > System > Setup > Access tab and adjusting the timeout value.
  5. Check Network Connectivity:
    • Ensure the cluster cable is properly connected and try using a different port if necessary.

If the issue persists, provide a sysdump file for further analysis. For more detailed guidance, consult the Exinda 7.5.6 Release Notes and consider contacting support for tailored assistance.

Frequently Asked Questions

What is a "split brain" scenario in Exinda clusters?
A "split brain" scenario occurs when both devices in a cluster show as master, leading to unstable connections and frequent logouts from the cluster IP.
How can I downgrade my Exinda devices to a previous version?
To downgrade, switch the boot partition to the one containing the desired version (e.g., 7.5.5) and reboot the appliance into that version.
What should I do if the cluster issue persists after troubleshooting?
If the issue persists, provide a sysdump file for further analysis and consider contacting support for more detailed assistance.
Choose files or drag and drop files
ai_initiated
atlas_studio
kb_automation
Was this article helpful?
Yes
No

  1. Priyanka Bhotika

  2. Posted

Comments