[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference azur::mcc

Title:DECmcc user notes file. Does not replace IPMT.
Notice:Use IPMT for problems. Newsletter location in note 6187
Moderator:TAEC::BEROUD
Created:Mon Aug 21 1989
Last Modified:Wed Jun 04 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:6497
Total number of notes:27359

3736.0. "snmp am questions?" by CGOOA::KUHNEN () Mon Sep 14 1992 20:44

    I am running v1.2 of DECmcc and am having some problems with the snmp
    access module.  
    
    From what I understand in the manual I should be able to monitor an
    a number of entities using a wild card. An example would be 
    snmp wellfleet interface * total utilization. When I do this I get an
    error stating that the wildcard cannot be used in this manner. Am I
    misunderstanding the manual or is there a problem with MCC ?
    
    When I monitor a number of entities individually after a period of time
    I start to get " software logic error
    detected<EOS>%MCC-S-SPECIALIZED_EXC, success with management
    module-specific exception reply" type errors. The alarm rules however
    are normally evaluated without a problem. These errors only occur
    when a large number of alarm rules are enabled.I have also seen
    "internet communication device error" messages under the same
    circumstances.
    
    Is there some tuning that must be done in MCC or UCX or VMS when a
    large number of SNMP type devices are to be managed.
    
    Thanks in advance for you help
    
    Harry Kuhnen
    
    
T.RTitleUserPersonal
Name
DateLines
3736.1YAHEY::BOSETue Sep 15 1992 19:0419
	Harry,

		The attribute Total Utilization is in the Statistics partition
	and handled by the DECmcc Performance Analyzer, which does not allow
	wildcarding of entities. However, wildcarding should work fine for
	requests on all other partitions, which are handled by the SNMP AM.

		The exceptions that you are seeing with large number of 
	alarm rules running are probably due to insufficient resources.
	The "internet communication device error" exception could be generated
	when there are two many sockets open. If you are using UCX on VMS
	you might be able to increase that maximum number of open sockets
	by modifying the following sysgen parameters :

		GBLPAGES = 1800
		NPAGEDYN = 342000
		NPAGEVIR = 342000
	Rahul.
3736.2parameters okayCGOOA::KUHNENWed Sep 16 1992 01:2413
    Rahul,
    
    	Thanks for the info on the performance analyzer. I missed that.
    My sysgen paramters are already much higher than you indicated for
    those values and I have in excess of 10000 free global pages. I have
    check all my sysgen values and account quotas against those in the
    release notes and they are all okay. I am running UCX v 2.0. would this
    have any bearing on the problem. I have also tried increasing the
    number of stacks available to snmp without any change in symptoms.
    
    Regards
    
    Harry