How to update a cluster
- 3 minutes to read
- Print
- DarkLight
- PDF
How to update a cluster
- 3 minutes to read
- Print
- DarkLight
- PDF
Article summary
Did you find this summary helpful?
Thank you for your feedback
Important
If you’re using a version prior to 3.33 and wish to update senhasegura to version 3.33 or later, please refer to the article How to update versions prior to 3.33.
Important
If you use Zabbix, it will be necessary to reconfigure it after updating senhasegura to version 3.30 via Orbit Command Line Interface (CLI). For instructions on how to configure Zabbix via Orbit CLI, refer to the article How to configure Zabbix via Orbit CLI.
In this article, you’ll learn how to update a senhasegura in a clustered environment.
Requirements
- Firewall Rules: Ensure your firewall settings allow communication with the senhasegura repositories. Detailed firewall rules can be found in the documentation.
- Cluster Synchronization: All cluster members must be synchronized.
- Server Time: Verify that the server's time is correctly set.
- (Situational) Arbitrator: If your environment includes an Arbitrator, remove it from the cluster.
- Snapshot: Take a system snapshot before proceeding with the update.
- Updating Sequentially: For versions that are at least two releases old, update sequentially as instructed in this article.
- New senhasegura Repository: Learn how to set up the new senhasegura repository.
Application unavailability during update
When senhasegura initiates the update process, it will be unavailable until the activity is complete.
Take a snapshot
Important
- In clustered environments, always take snapshots in reverse order, starting from the last member towards the primary member.
- Always take snapshots of all senhasegura instances separately.
- Turn off one instance at a time, take the snapshot, turn it back on, and validate that it has resynchronized before proceeding to the next instance.
Snapshots should be taken only when senhasegura instances are offline. This ensures the ability to revert to a previous version if necessary.
- To shut down a senhasegura instance, use the following command:
sudo orbit shutdown
2.Once the instance is completely shut down, take a snapshot in the hypervisor.
3. Then, start the senhasegura instance and update the application after taking snapshots of all instances in the cluster.
How to update a cluster
Caution
Before updating senhasegura, always take a snapshot in your hypervisor and backup using the Orbit CLI.
Caution
If you use the senhasegura Arbitrator, remove it from the cluster before updating senhasegura. Learn more in our documentation.
1.In the senhasegura platform, go to Orbit Config Manager > Replication > Status and check the status of each member. All members must be ON and synchronized.
- In the senhasegura platform, go to Settings > Execution processes, and turn the following items OFF:
- Expired password.
- Expired operation cancellation.
- Operation executor.
- Operation executor: Retry.
- Take snapshots of each cluster member.
Important
Remember to take snapshots in reverse order, from the last member to the first.
Start an SSH session as an administrator user of senhasegura.
Run the following command:
apt-get update
- Run the following command:
apt-get install orbit-cli
- Run the following command to update senhasegura to the latest version:
orbit update
- After updating senhasegura, restart the system to ensure all changes have been applied correctly:
orbit shutdown -r
Important
When updating Senhasegura in a clustered environment, do not update the members as standalone entities. Be sure to update each cluster member individually, starting with the primary member and proceeding sequentially to the other members. For example, update the Primary Member, then update Member 2, and finally, update Member 3.
Not following these steps during the update can harm the cluster.
- Apply steps 5 to 8 to all the other cluster members.
- Go back to Settings > Execution processes and re-enable the items:
- Expired password.
- Expired operation cancellation.
- Operation executor.
- Operation executor: Retry.
- Return to Orbit Config Manager > Replication > Status, and check the status of each member to confirm that they are ON and synchronized.
Do you still have questions? Reach out to the senhasegura Community.
Was this article helpful?
