When a test alarm is triggered from the robot it shows a correct server name in the IM Alarm sub-console, but when the same alert is transferred to SOI the server or source name appears incorrect.
There could be a mixup of the CI causing this.
1. Check the results of this query:
select * from CM_COMPUTER_SYSTEM where name = '<system_name>'
2. From the Operator Console, please follow the steps below:
Select <system_name> in the Inventory and then under Actions-> Delete Selected
DESELECT the three options noted below so that none are selected.
- Prevent rediscovery
- Acknowledge alarms
- Delete measurements
3. RDP to the robot itself and clear the niscache folder of all data
How to clear niscache on an individual robot
- Access Infrastructure Manager
- Navigate to the robot, select the controller probe, and press Ctrl-P
- The probe utility dialog opens
- Click the Options icon
- Select the Expert Mode option, and click OK
- In Probe Command set, select _nis_cache_clean, and click the arrow (green) to send the command request
- In Probe Command set, select _reset_device_id_and_restart, and click the arrow (green) to send the command request
- The niscache is now clean and the robot is restarting
4. Run the Operator Console discovery wizard to rediscover the system.
If rediscovery does not work, then refer to the techdocs page below for enabling this option in the Connector configuration file.
Additionally, make sure to check for any database errors, e.g., communication link errors, or Microsoft SQL Server Native Client 11.0 errors.
If databse is in a cluster configuration, also check for any cluster-related errors or data_engine timeouts due to communication link errors.