Hello guys,we have a problem with a management sever at a customer site:We did not see pending agents on this MS on monday and rebootet the machine. After that the MS came up without any errors but:- it does not monitor any agents- it does not monitor himself- just produced 21023 every 5 Mins. and every few hours an 7019The environment:- all MS SCOM 2007 R2- clustered RMS- 4 MS, 2 GW MSWhat we did so far:- Checked the eventlog of the RMS: no errors, when the faulting MS is starting, no errors, after the MS was started- rebootet the MS, cleaned the HS cache- created a complete Trace, but no errors could be found. The trace always says, that the RMS connection could be successfully establishedAny hints from you guys, what we could check to find the root cause for this problem?Any suggestions will be highly appreciated :)Best regards,Dirk

Hi,Could you post that event?Anders Bengtsson | Microsoft MVP - Operations Manager | http://www.contoso.se

Sure:Event 21023: OpsMgr has no configuration for Management group XXX and is requesting new configuration from the Configiration Service.

Hi Dirk.Let's start simple here. Is the MS capable of finding the IP address of the RMS based on its FQDN? It could be a DNS problem.Also, in the SCOM Console, does this MS show up as a Management Server? Does the OpsMgr eventlog on the RMS tell you anything more?Has the name of the MS been changed after SCOM has been installed? Can the RMS find the MS based on tis FQDN?Best regards, Marnix Wolf (Thoughts on OpsMgr)

Hi Marnix,nothing has changed since setup 4 month ago.IP, DNS etc. is all ok, because the MS can establish a session with the RMS, no name change. MS shows up as a MS, MS State is healthy from HSW and Server side.Regards,Dirk

Not a single event on the RMS side when the MS has its health service restarted?Best regards, Marnix Wolf (Thoughts on OpsMgr)

When i restart the HS on the MS i get an Event 29103 "HS running on MS XXX has contacted OpsMgr config service to synch... State cookie is ..." That's all. No errors, just "good confirmations" in my point of view...So communication between MS and RMS should be ok.

Hmmm.Let me search my archives. Have had an almost similar issue before. I'll get back to you asap.Best regards, Marnix Wolf (Thoughts on OpsMgr)

We had the same problem. It is probably your RMS that is preventing the updates. You should stop your RMS (take the cluster resources offline andthen bring only the shared disk online), clear the Health Service State on your RMS and then recycle all SCOM services on your RMS (bring the cluster resources online).Updates will then be sent to your MS and monitoring will start again. We now check hourly on the modification dates of the configuration files. If they differ to much we create an alert. Regards,Marc Klaverhttp://jama00.wordpress.com/

Hi Marc,your are a genious!!! :)That solved our problem!!! After resetting the HS cache on the RMS the config gets updated on the MS. I really forgot to check this possible error source...Many thanks,Dirk

HiMarc beat me to it. Good job Marc! :)Best regards, Marnix Wolf (Thoughts on OpsMgr)

Hi Marc,your are a genious!!! :)Dirk grrr never call marc a genious. i have to hear that for at least a week now :)But seriously, we haven't had this problem since we totally excluded virusscanning from all directories of scom (we did have exclusions, but not for highrisk processes). Most likely the problem is a filelock on a EDB, CHK orLOG filewhen configuration is updated (as simple as approving a server or setting an override).http://blogs.technet.com/operationsmgr/archive/2008/11/05/recommendations-for-antivirus-exclusions-in-mom-2005-and-opsmgr-2007.aspxRob Korving http://jama00.wordpress.com/

Hello all,Over the last 12 weeks we have experienced similar "orphaning" of some of our Management Servers as mentioned above and we too have used the clearing of the RMS Health Cache (by deleting the Health Service State directory after stopping the Health Service on the RMS) to get the Management Server active again. It can also affect Gateway servers as they depend on the Management Server to forward their configuration. Once you reset the Health Cache on the RMS the Gateway becomes alive again as soon as the Management Server sends the update. An easy way to spot one of these "orphaned" Management servers is if it takes a very long time (like never) to discover new Agents - if you get this, go check for the 21023 event on the MS to confirm. If you see 21023 repeated every 5 minutes and it is the only event, you have the above conditon.This even happens after applying the latest CU1 for R2 release.I have now excluded the whole SCOM dierctory (typically C:\Program Files\System Center Operations Manager 2007) from being scanned by our AV and will report back if this solves our problem. We run a really large distributed environment and we see this "oprhaning" of a Management Server at least 3-4 times per month, so it should be relatively easy to judge if this has fixed it... thank you for the suggestion. There are some consequences to clearing this cache, so it is not something we would want to do every week... so hopefully we have a solution here.I'll keep everybody posted.RegardsChristo

kindly connect to my problem and help us to resolve the same .. http://social.technet.microsoft.com/Forums/en-US/operationsmanagerdeployment/thread/b58f8860-3a19-43f2-a7f3-69b199dfcf36/#25a8f3e3-2acc-4dae-a227-5dfdfad9946b thanks