VMWareProbe failing: Unable to establish connection to https://<IP>/sdk

cloudyrobert
Kilo Guru

Anyone solved this Discovery issue? Historically, we've successfully discovered our vCenters, but it's stopped working relatively recently. We're getting:

Unable to establish connection to https://<IP>/sdk

We attempted the guidance in this KB under 'Probe "VMWare - vCenter Datacenters" fails with "Unable to establish connection to https:///sdk"'. We're able to "successfully" hit /sdk from the browser on the MID host. I put that in quotes, because it's not a web server, so you don't really get a page. We're able to successfully log into /mob using the same domain credentials that we're using with Discovery.

Edit 1:

Looked at this too. The payload shows that it used the correct VMWare credential, and the cred is pre-fixed with the domain.

Edit 2:

We're seeing the error "SEVERE *** ERROR *** An error occurred while decrypting credentials from instance" in our MID logs. This isn't OAuth, but wonder if this is relevant. We're patching to New York Patch 8 in less than a week, but we're not there yet.

Edit 3:

We just ran into this orphaned record issue with the connection table related to an integration after a clone due to the preservers only being set up for the child tables, but I don't think that's the issue here. I can open the credential records as I would expect.

Edit 4:

Just in case... I deleted the keystore & re-validated the MID server.

1 ACCEPTED SOLUTION

cloudyrobert
Kilo Guru
10 REPLIES 10

So I reply to myself. 
Although it was already done, we came again over setting  'Applies to: Specific MID servers' on the credential record, matching the MID server specified in the Discovery Schedule and it worked.

I don't have the confirmation but I guess this happened because the credential was manually added, but the Discovery Schedule was created using an Update Set, and because it has a specific MID Server defined something related to the sysid may have conflicted.

Anyway, @cloudyrobert , @Ashutosh Munot1  and @HugoMeeuwes thanks for all the information in this post as it helped us to troubleshoot systematically and I understood some key concepts.