EDR source server not converging
Hi,
We have a server that automatically reboots every day at 7am and is also replicating from GCP to AWS via Elastic Disaster Recovery. The source server we created never goes into the ready state permanently, but has been fluctuating between initial sync and ready and constantly rescanning. After several days, we see that it has finally gone into the "not ready" state, with the error "not converging". We have verified that the disk settings are correct. Is it possible that the daily restart of the replication agent is causing the failure?
Thanks.
- Newest
- Most votes
- Most comments
Based on the information provided, it's possible that the daily restart of your server at 7am could be interfering with the replication process in AWS Elastic Disaster Recovery (EDR). The constant fluctuation between initial sync, ready, and rescanning states, followed by the "not ready" state with a "not converging" error, suggests that the replication process is being disrupted regularly.
The replication status of a source server in EDR is crucial for ensuring that it's ready for recovery. Normally, a healthy replication process would show a "Ready" status under the "Ready for recovery" column and a "Healthy" status under the "Data replication status" column.
The daily reboot of your server at 7am could be causing the replication agent to restart, which might interrupt the ongoing replication process. This interruption could explain why the server is not able to maintain a stable "Ready" state and keeps reverting to "Initial sync" or "Rescanning".
To troubleshoot this issue, you could try the following:
- Temporarily disable the daily reboot and observe if the replication status stabilizes.
- If possible, schedule the reboot for a time when data changes are minimal to allow the replication process to catch up.
- Check the logs of the replication agent to see if there are any error messages coinciding with the reboot time.
- Ensure that the replication agent is configured to start automatically after a reboot.
- Verify that there are no network issues or bandwidth constraints that might be exacerbated by the daily reboot.
If the problem persists after trying these steps, you may need to contact AWS support for further assistance. They can help investigate if there are any underlying issues with the replication process or provide guidance on optimizing the replication configuration for your specific use case.
Remember, a stable and healthy replication status is crucial for ensuring that your source server is ready for recovery when needed.
Sources
Performing a failover with Elastic Disaster Recovery - AWS Elastic Disaster Recovery
Interacting with the Source Servers page - AWS Elastic Disaster Recovery
answered 2 years ago
The daily reboot of the source server can interrupt the AWS Replication Agent, which causes replication to fail and prevents it from converging. During reboot, the agent stops, and if a sync is ongoing, it will restart or fail, resulting in repeated sync attempts and eventually a "not converging" state.
Temporarily disable the daily reboot to allow DRS to complete the initial sync without interruption.
Check replication agent logs for interruptions:
Linux: /var/log/awsdr/
Windows: C:\Program Files (x86)\AWS Replication Agent\logs\
Ensure you're using the latest agent version: AWS DRS – Installing/Updating the Replication Agent
https://docs.aws.amazon.com/drs/latest/userguide/troubleshooting.html#troubleshooting-not-converging
answered a year ago
