I've just installed UDT 3.2 into my existing Orion system. I am running NPM 11.0.1, SAM 6.1.1, NTA 3.11. I have added my 10 Domain Controllers in UDT, but 3 of them come back with the following status, "Polling Failed, check security log access credentials" The domain controllers are 2008R2, 2012 and 2003. The account I am using is a member of the Domain Admins group. I've checked every permission I can think of on the logs, but the account has all needed permissions. I've been able to remotely access the logs with the account. I'm not sure what else to check as I can't find any other errors. When I tested UDT 3.1, I did not have this issue. I am using the same account as my test. Thanks for any pointers.
No, I did not open a ticket. That was gong to be my next step, but I've found a workaround.
To bring up to date - I have been operating under the impression (determined early on in this thread) that this error was a false positive. I discovered a couple of days ago that this is not the case. When I have a DC in a failed poll state, it is not collecting login information. I had 2 DCs that had been in a failed poll state for 2 weeks. I verified the last logon record for that domain in the Orion database was 2 weeks old.
Workaround - In the past, I've been deleting and then re-creating the node. Then I found something that's been in front of my face the whole time. There is a way to manually force a poll. Go to Settings --> UDT Settings --> View UDT Job Status. Find the failed DC and click the Poll Now button. Immediate relief.
Some History - In earlier versions of UDT when you went to UDT Settings --> Manage Active Directory Domain Controller, it automatically forced a poll of all managed DCs. I found some entries in this forum complaining that this was a real pain in cases where many DCs were managed. There were requests to stop this behavior and replace it with a manual alternative. Apparently the Orion/UDT team was listening and this is the result. So on one side this is a good thing, but obviously the scheduled polling shouldn't be failing and must be addressed. I'm sure they're monitoring this thread, but opening a ticket may be the best way to ensure a fix is in the works.
Sometimes the AD credentials test running with each poll behaves weirdly and so I would like to ask: are the respective data polled? (despite the error..) It easily could be kind of "false positive" alert with data still being correctly polled...
I think you are correct. Data does appear to be collected. Of course, not every DC in a domain is failing. At some point after synchronization, that data could be coming from another DC. Regardless, it's annoying at the very least. So far, deleting the node and then re-adding it has resolved the issue. I know this may not be an option for many, but that's the road I'm taking. If a node that I have already re-added starts failing again, I'll report back.
Thanks for the suggestion, but It didn't work too well for me. I deleted the 3 DCs that are giving the error and added them back. Things were fine until I came in this morning. 1 of the 3 is back to being in an error state and now I have 2 new ones that are showing the error. I'm going to delete them and see what happens.
This is happening to us as well. We are monitoring 12 DCs. Shortly after upgrading UDT to 3.2 a couple of weeks ago one of them began getting the "Polling Failed" error. I lived with it for a few days while I researched possible solutions. In the end, I deleted the node and then added it back. Since then, there has been no issue with that DC, but 2 others have now started to fail. We were running NPM 11.0.1, NTA 4.1 and UDT 3.1 prior to the upgrade and had been monitoring these DCs for several months without issue. UDT is the only module we upgraded.
SolarWinds solutions are rooted in our deep connection to our user base in the THWACK® online community. More than 150,000 members are here to solve problems, share technology and best practices, and directly contribute to our product development process.