So I just installed SupportAssist and everything worked great... except on my computers with multiple IP addresses. I found this little gem in the Troubleshooting section:
The collection may fail on a Windows server if the server has multiple IP addresses or the first IPv4 address of the server populated by the Dell Windows Server Management Pack is not configured correctly. To ensure that the collection tool (DSET) runs on the Windows server, configure the first IPv4 address of the Windows server in Operations Manager
But I don't know how to configure the first IPV4 address in Operations Manager. I have scoured their documentation and only found mentions of the bind order. So I changed the bind order of that server so that the IP I want is first... and it still shows in support assist as the non-routable iSCSI ip. Any ideas?
This is a known big with SCOM SupportAssist and will be fixed in next release(probably sometime early next year). we have filed a defect in JIRA SASC-242 to track the fix.
Here is the email thread w.r.t this issue:
SupportAssist (SCOM) log collection is failing for the managed node configured with multiple IP's
Defect reported by TAM
One of our customer is facing issue with collection failure in DSA (v1.1) for SCOM.
Following is the setup:
The host has two NIC’s, one is configured with 10.* and the other with 169.*
169.* is for internal communication.
10.* is connected to the network.
Now the SCOM agent by the looks of things are having the 169.* as the first IP in the agent properties where for other hosts it has the 10.* as the first IP. Looks like support assist configures itself with the first ip of the SCOM agent.
Is there any way that we can change this in support assist?
As per inputs received from development team, this is the current design where we store the list of IP for each client(managed node)in a table, the IP list is received from SCOM and we don't have control on the order.
SCOM Plugin picks up the first IP from the table and tries for a collection. We will need to change the logic at plugin side where we check each and every IP to see if its reachable if the first IP fails for log collection.
As per current design, we check only the first IP returned by SCOM and try for a collection. If it fails due to connectivity issue, we dont recheck with next IP in the list.