[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

1840.0. "Alarm Error Accessing Entity" by BABY::MOTYL () Thu Nov 21 1991 16:59

Listed below is an alarm notifaction and the alarm rule.  Could someone please
help me identify why I keep getting an alarm notifaction when there is no 
problem.  This doesn't just happen to one specific rule, but sporadically to 
all of them, and each time I get the same exception message:

	Exception:     Error occurred while accessing the entity.
                                                          	
Have searched through the the notes file and the documentation, but found 
nothing specific regarding this problem.

Thanks,

Joe 

*********************************************************************************
From:	BABY::SYSTEM       "WR3DR1 - IVODR1 CIRCUIT" 19-NOV-1991 07:48:50.52
To:	 @SYS$MANAGER:MCC.DIS
CC:	
Subj:	Notification of exception " MCC 0 ALARMS RULE WR3DR1_SYN2  19-NOV-1991 07:48:17.62"

    Rule name:     MCC 0 ALARMS RULE WR3DR1_SYN2
    Domain:        Domain BABY_NS:.WR3
  Occurred at:     19-NOV-1991 07:48:17.62
     Category:     ***** Circuit Substate has changed *****
  Description:     WR3DR1 - IVODR1 CIRCUIT PROBLEM
     Severity:     critical
 
   Expression:     (change_of (node4 WR3DR1 circuit syn-2 state,*,*), at every=00:30:00)
   Exception:     Error occurred while accessing the entity.

*********************************************************************************

Create MCC 0 ALARMS RULE WR3DR1_SYN2  -
  Category          = "***** Circuit Substate has changed *****", -
  Description       = " WR3DR1 - IVODR1 CIRCUIT PROBLEM ", -
  Expression        = (change_of (node4 WR3DR1 circuit syn-2 state,*,*), at every=00:30:00), -
  Procedure         = MCC_COMMON:MCC_ALARMS_MAIL_ALARM.COM     , -
  EXCEPTION HANDLER = MCC_COMMON:MCC_ALARMS_MAIL_EXCEPTION.COM , -
  Parameter         = "@SYS$MANAGER:MCC.DIS", -
  Queue             = "BABY$BATCH", -
  Perceived Severity = CRITICAL, -
  In Domain         = WR3
*********************************************************************************
T.RTitleUserPersonal
Name		DateLines
1840.1polling prob??JETSAM::WOODCOCKThu Nov 21 1991 18:1821
>>	Exception:     Error occurred while accessing the entity.
                                                          	
Hi Joe,

The experience I have with the above error is typically that the poll failed.
It could be because the node only supports single links (DR250,DECSA), the
node is overloaded with traffic (mngmt messages are the first to get dropped), 
your system is overloaded and can't handle the poll rate, or you've got network
problems getting to the node. Pick one. If this is a 100% consistent error
try a SHOW command for the same info for a sanity check. It could also be 
something to do with accounts or privs at the node.

>>  Expression        = (change_of (node4 WR3DR1 circuit syn-2 state,*,*), at every=00:30:00), -

Also, you probably want to monitor the 'substate' rather than 'state' if 
you're looking for circuit up/down indicators. State shows whether the circuit
is turned ON or OFF (characteristic) while substate shows up or down (status).


best regards,
brad...
1840.2BABY::MOTYLFri Nov 22 1991 11:578
    
    Let me tell you a little bit about this network.  The entire network is
    made up of 56kbs circuits countected though dec router 2000, and these
    routers are hosted by 3500's.  As of lately the utilization on this
    network has been low due to the current status of the ACT's, so not
    being able to poll doesn't seem to make sense.  As far as accounts or
    privs, could you be a little more specific.
1840.3MCC needs much resourcesMFRNW1::SCHUSTERKarl Schuster @MFR Network ServicesFri Nov 22 1991 13:2212
    I get the same error relativly often for slow or busy DECnet nodes.
    I have tuned the systems ( Pool, MAXPROCESSCOUNT, Quotas ...), but
    we still get "error accessing entity" errors.
    NCP connections work fine.
    
    The reason is, that MCC really needs much resources on the remote
    systems to be polled ( e.g. multiple management links ).
    
    Perhaps there will be changes in V1.2 ?
    
    Karl
1840.4BABY::MOTYLMon Nov 25 1991 17:569
    Karl, in an effort to save some time, could you elaborate some on the 
    specfice tuning that you did, also, is this tuning talked about in any
    of the alarms docs, if so where? 
    
    Thanks
    
    Joe
1840.5We still get the errors ....MFRNW1::SCHUSTERKarl Schuster @MFR Network ServicesTue Nov 26 1991 05:529
    re .4
    I did not do great things.
    You should consider the following (for the remote system):
    - enough process entry slots (MAXPROCESSCNT)
    - enough pool space ( 25% free ) SRPcount,IRPcount,LRPcount
    If the CPU is overloaded, then you have no chance.
    In many cases we still get the errors, and I hope that MCC V1.2
    consumes less resources for polling.
    Karl