cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
695
Views
15
Helpful
14
Replies

DFM 2.0.6: All devices are in learning state

peteroseneff
Level 1
Level 1

Hello!

Discovery Status of all 269 devices is Learning.

I reinstall the whole DFM module with updates in my LMS system with no result.

I suspect the situation above appeared due to PTMServer process problem. When I try to check DFM Polling Parameters in the Configuration --> Polling and Thresholds page I get the following error message:

##############################################

"Could not retrieve current data since the PTMServer process is down.

Please restart the process and perform this operation again"

##############################################

The PTMServer process is running. I restart the PTMServer process and get the same error message.

Is there any influence of PTMServer process to DFM devices discovery?

The PTMServer.log shows the following:

##############################################

27/Dec/2006 16:58:56:603 ERROR main com.cisco.nm.xms.ctm.client.CTMCall establishIPC - URN_NOT_FOUND : urn "VPTA" : Not found !!

27/Dec/2006 16:58:56:618 ERROR main com.cisco.nm.xms.ctm.client.CTMCall establishIPC - URN_NOT_FOUND : urn "VPTA" : Not found !!

Wed Dec 27 16:58:56 MSK 2006 PTAProxy :: CTMException in PTASyncUp(delGrpFlag) :: com.cisco.nm.xms.ctm.common.CTMException: URN_NOT_FOUND : urn "VPTA" : Not found !!

##############################################

As far as I understand it is not critical error?

Why the DFM is unable to communicate with the PTMSever process?

Thank you for your help!

P.S. LMS 2.6 is running on Windows 2000 Server SP4 with 2GB of RAM

14 Replies 14

Joe Clarke
Cisco Employee
Cisco Employee

The PTMServer problem is a symptom of a larger problem. I'll bet all of your devices are hanging at 10%. The reason is that sm_server (i.e. DfmServer) is having a problem. Check the NMSROOT\objects\smarts\local\logs\DFM.log for relevant errors. Also, make sure your server's COMPUTERNAME AND fully-qualified hostname properly resolve. Use the resolver.pl tool to confirm this. For example, if your COMPUTERNAME is cworks, and your fully-qualified hostname (as seen in your computer's identification tab) is cworks.company.com:

NMSROOT\bin\perl NMSROOT\bin\resolver.pl cworks

NMSROOT\bin\perl NMSROOT\bin\resolver.pl cworks.company.com

Both commands should return the same IP address.

Sir,

I have the following entries in the DFM.log:

[27-Dec-2006 4:03:31 PM+569ms]

RSRV-N-DMSTARTED-Domain Manager DFM started on lms:39614

V2.0 IDU6(63983), 20-Jun-2006 11:19:58 - SP 6, Build 3

ICF-N-SAVESTART-PersistenceManager: save started at 27-Dec-2006 10:03:32 PM

ICF-N-SAVEFINISH-PersistenceManager: save finished at 27-Dec-2006 10:03:32 PM

CMF_ProcessManager::ProcessNextMessage() -- Non-ShutdownRequest

CMF_ProcessManager::ProcessNextMessage() -- Non-ShutdownRequest

CMF_ProcessManager::ProcessNextMessage() -- Non-ShutdownRequest

CMF_ProcessManager::ProcessNextMessage() -- Non-ShutdownRequest

CMF_ProcessManager::ProcessNextMessage() -- Non-ShutdownRequest

CMF_ProcessManager::ProcessNextMessage() -- Non-ShutdownRequest

CMF_ProcessManager::ProcessNextMessage() -- Non-ShutdownRequest

CMF_ProcessManager::ProcessNextMessage() -- Non-ShutdownRequest

CMF_ProcessManager::ProcessNextMessage() -- Non-ShutdownRequest

ICF-N-SAVESTART-PersistenceManager: save started at 28-Dec-2006 4:03:32 AM

ICF-N-SAVEFINISH-PersistenceManager: save finished at 28-Dec-2006 4:03:32 AM

CMF_ProcessManager::ProcessNextMessage() -- Non-ShutdownRequest

ICF-N-SAVESTART-PersistenceManager: save started at 28-Dec-2006 10:03:32 AM

ICF-N-SAVEFINISH-PersistenceManager: save finished at 28-Dec-2006 10:03:32 AM

CMF_ProcessManager::ProcessNextMessage() -- Non-ShutdownRequest

CMF_ProcessManager::ProcessNextMessage() -- Non-ShutdownRequest

CMF_ProcessManager::ProcessNextMessage() -- Non-ShutdownRequest

I'm unable to decode it. What can you say?

P.S. COMPUTERNAME and FQDN resolve OK.

This log does not indicate an obvious error with DfmServer, but if the VPTA URN cannot be found, then there is some problem communicating with it. What does the full pdshow look like?

Sir,

The pdshow command output inside the attachment.

pdshow looks okay. I see nothing obvious that would explain why the VPTA URN is not registering. I recommend you open a TAC Service Request so that they can collection additional logs to analyze this problem further.

Understood. Thank you.

jefflevyadc
Level 1
Level 1

I have the same exact issue--same errors and symptoms. I have a case open on this. Did you get any ideas on this yet?

If you see the error that VPTA is not found, that always points to a problem with the DfmServer process. You should focus on the following things:

* Make sure both DfmBroker and DfmServer are running

* Check to make sure the right number of sm_server and brstart processes are running

Solaris should have one of each Windows should have two of each

* Make sure tcp/9002 is not in use by another application (other than brstart)

* Make sure the server?s fully-qualified hostname is resolvable

If that looks good, open a TAC Service Request with the NMSROOT/objects/smarts/local/logs/DFM.log and NMSROOT/log/dfmLogs/TIS/* files.

The DFMserver process won't stay running. I can restart it and it just stops again. I attached the DFM.log file. I don't have a TIS directory in the location you list.

There could be a problem with your NMSROOT/objects/smarts/conf/clientConnect.conf/serverConnect.conf files, but those messages are from yesterday, so my thought is that sm_server is not even trying to come up. What does the output of pdshow look like? On what platform is LMS installed?

You should definitely have a TIS directory in the location I listed unless someone purged the NMSROOT/log directory (which would be very bad). what directories do you have under NMSROOT/log/dfmLogs?

This is on Windows2003 SP1.

In the dfmlogs directory: AAD,cfi,DDV,DPS is all.

Attached is the pdshow output.

I don't think we've ever purged the log directory.

The missing log directories is definitely problematic. You should try uninstalling, then reinstalling DFM first, then see if the DfmServer process is up and running. Also, make sure when you should down crmdmgtd that you do not see any sm_server processes running.

Of course I meant, "when you SHUT down crmdmgtd ..."

Dear jclarke:

We had some trouble with LMS 2.6, and it`s no easy to run through the whole re-installation (2.5, 2.5.1 and 2.6)

Now, it looks as if almost everything is working fine, except DFM 2.06.

I`m suffering this same sympthom.

I cannot delete devices (stay in pending state forever) nor rediscover devices (they stay in learning state forever).

What could I do before thinking of re-installing everything again?

E:\CTH>E:\"Program Files"\CSCOpx\bin\perl "\Program Files\CSCOpx\bin\resolver.pl

" rcdo-netmgmt

Original name: rcdo-netmgmt

Hostnames/IP Addresses:

rcdo-netmgmt/192.168.242.10

E:\CTH>luk rcdo-netmgmt.rcdo

E:\CTH>E:\"Program Files"\CSCOpx\bin\perl "\Program Files\CSCOpx\bin\resolver.pl

" rcdo-netmgmt.rcdo

Original name: rcdo-netmgmt.rcdo

Hostnames/IP Addresses:

rcdo-netmgmt.rcdo/192.168.242.10

Thanks a lot

Julio

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: