Installing a mirror image disaster recovery environment, but the DA keeps shutting down.
Article ID: 439868
Updated On:
Products
Issue/Introduction
New Data Aggregator install instance not starting (getting heartbeats from somewhere else)
I installed a new DA 25.4.5 instance on the Disaster Recovery site (*.*.*.134),
DA fails to start complaining about heartbeat from the secondary (main) DA.
[root@host:/opt/NetOps/IMDataAggregator]# cat shutdown.log
2026-05-07T09:28:00,035 | INFO | main | shutdown | ase.heartbeat.DBStateManagerImpl 1130 | - - | | Local addresses found: [*.*.*.134]
2026-05-07T09:28:00,932 | ERROR | main | shutdown | ase.heartbeat.DBStateManagerImpl 1249 | - - | | Shutting down DA due to DB contention. Expecting DB heartbeats from only source DA *.*.*.134, but saw heartbeats from DAs: *.*.*.129
*.*.*.129/.130 pair is installed on a primary site.
Environment
DX NetOps 25.4.5
mirror image disaster recovery (DR) lab
Cause
We checked the primary env, and found that the dbconnection.cfg files on the DA’s had one vertica host from the secondary environment listed.
Resolution
1. Correct Configuration Files
Manually edit the dbconnection.cfg files on the Primary Site Data Aggregators. Ensure the configuration includes only the primary Vertica hosts. Remove any IP addresses or hostnames belonging to the secondary (DR) environment.
2. Service Management
- Stop the DA services on both Primary Site Data Aggregators.
- Perform a health check on the Primary Vertica database by running the
etlhealthscript to ensure there is no residual contention or lock-related issues.
3. Restoration of Services
- Once the configuration is corrected and the database health is verified, restart the primary DA services.
- Set both Primary DAs to Active status.
- Proceed with starting the DA instance at the DR site
Feedback
