Sorry this took me a while to back to you. I tried your advice and it worked for a while so I've been doing some troubleshooting and here is what I found:
we have have an active/active sql failover cluster with multiple instances..
- Add the VIP on the first instance to Insight for sql.
- allow the initial poll to complete.
- add the second instance
- wait for the poll to complete on the second instance.
- both instances now show time out on the poll scheduler.
- remove the second instance from insight.
- Unmanage the first instance.
- re-manage the first instance.
- Poll completes properly.
It seems that it won't work with more than one instance on a cluster?
Any ideas?