-
Title
Agent goes offline when metadata refresh is initiated -
Description
Specific agent machine is online (green), but getting no disks information. When metadata refresh is initiated, agent goes offline. Running web test https://<agent_name>:8006/apprecovery/api/agent/pair/connect works every time. -
Cause
Multiple agents with same host URL but different Agent IDs -
Resolution
Workaround:
The issue has been traced to the Core server, having multiple entries for the same agent machine. Different entries are having different Agent IDs, and only 1 ID is matching with the ID from the actual agent's registry.
Navigate to registry on the Core into HKLM\Software\AppRecovery\Core\Agents\ and check AgentIDs one by one, writing down entries with the same agent machine name or IP. Write down display names.
Check these display names in Core console and see if there are Recovery Points, that belong to them.
On the agent machine, open regedit, HKLM\Software\Apprecovery\Agent\AgentID and write down agent's ID
Back to the Core, remove all display names that having same name/IP for the URL, and whose agent ID doesn't match. Remaining machine with the proper name or IP and matching AgentID should be online. Removed agents' RPs will be moved to the Recovery Points only category.
-
Additional Information
When agent is starting, Core sees it as green because URL is responding. When metadata refresh is initiated, agent ID is failing comparison and agent is being marked offline.