[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

1583.0. "ETHERNET_AM STATION alarm rule ACCVIO" by CUJO::HILL (Dan Hill-Net.Mgt.-Customer Resident) Thu Oct 03 1991 02:09

    I was attempting to create the following alarm rule, but an access
    violation resulted.  Any ideas?  Will one of the patches in note 1297
    solve this problem?
    
    $! CREATE_ETH_ALARM_RULE.COM
    $!    ETHERNET Station Down Alarm Rule
    $!
    $ MANAGE/ENTERPRISE
    CREATE MCC 0 ALARMS RULE KANG_STATION_DOWN -
      EXPRESSION = (STATION .station.KANG ALL STATUS ,-
    		    WITH RESPONDING ADDRESS <> AA-00-04-00-07-0C ,-
    		    AT EVERY 00:01:00) ,-
      PROCEDURE         = MCC_COMMON:MCC_ALARMS_LOG_ALARM.COM ,-
      EXCEPTION HANDLER = MCC_COMMON:MCC_ALARMS_LOG_EXCEPTION.COM ,-
      CATEGORY          = "Node down" ,-
      DESCRIPTION       = "Node is not responding." ,-
      QUEUE             = "MCC_ALARMS_BARCH" ,-
      PERCEIVED SEVERITY = CRITICAL ,-
      IN DOMAIN = .VMON
                     EXIT
    $ EXIT
    
    ------------
    
    $ @CREATE_ETH_ALARM_RULE.COM
    
    DECmcc (V1.1.0)
                                               
    %SYSTEM-F-ACCVIO, access violation, reason mask=00, virtual
    address=00000000, PC=0043FA58, PSL=03C00005
    %SYSTEM-F-ACCVIO, access violation, reason mask=00, virtual
    address=0000000A, PC=0000000A, PSL=0000000F
    %DCL-W-SKPDAT, image data (records not beginning with "$") ignored
                                               
T.RTitleUserPersonal
Name
DateLines
1583.1CorrectionCUJO::HILLDan Hill-Net.Mgt.-Customer ResidentThu Oct 03 1991 02:152
    Correction.  I meant the patches in note 1267.
    
1583.2Need more information.CHRISB::BRIENENDECmcc Bridge|Station|SNMP Management.Thu Oct 03 1991 11:5012
Hi Dan,

We need more information before your question can be answered.

Specifically:

 1. What is .station.KANG (e.g. DECconcentrator 500) ?
 2. Does SHOW STATION .station.KANG ALL STATUS work?
 3. What about the above command, with the "WITH RESPONDING ADDRESS"
     portion added?

						Chris
1583.3You can't put the WITH clause in a Rule ExpressionNANOVX::ROBERTSKeith Roberts - DECmcc Toolkit TeamThu Oct 03 1991 12:2719
RE: .0

>    CREATE MCC 0 ALARMS RULE KANG_STATION_DOWN -
>      EXPRESSION = (STATION .station.KANG ALL STATUS ,-
>    		    WITH RESPONDING ADDRESS <> AA-00-04-00-07-0C ,-
>    		    AT EVERY 00:01:00) ,-
>      PROCEDURE         = MCC_COMMON:MCC_ALARMS_LOG_ALARM.COM ,-
>      EXCEPTION HANDLER = MCC_COMMON:MCC_ALARMS_LOG_EXCEPTION.COM ,-
>      CATEGORY          = "Node down" ,-
>      DESCRIPTION       = "Node is not responding." ,-
>      QUEUE             = "MCC_ALARMS_BARCH" ,-
>      PERCEIVED SEVERITY = CRITICAL ,-
>      IN DOMAIN = .VMON

Unfortunatly, you cannot put a with clause in your Rule Expression.
I imagine this is what is causing the ACCVIO.  I don't know if this
has been QAR'd, but will check it out.

/Keith
1583.4More info on ETHERNET AM ruleCUJO::HILLDan Hill-Net.Mgt.-Customer ResidentThu Oct 03 1991 13:2114
    Re:.2
    
    .station.KANG is a DECnet Phase IV node
    SHOW STATION .station.KANG ALL STATUS works fine
    
    Re:.3
    I don't really care what the syntax is or what attribute I must
    evaluate.  All I am really trying to test is whether or not I can set
    up an alarm to tell me when a node goes down.  I want to use the
    ETHERNET_AM because it is faster than the DNA4_AM.  I plan to do the
    same for bridges and terminal servers.  This was just the first test
    rule I tried.
    
    Any ideas??
1583.5don't forget the occurs functionTOOK::CALLANDERMCC = My Constant CompanionThu Oct 17 1991 20:107
you can also wait on events as well as monitoring attributes. Along
with that you can make use of the "exception condition" for detecting
a problem (when a node is unreachable it simply doesn't reponsd, so you
test for something always true and when it is false yiou have lost 
contact).

jill
1583.6halted nodes can still return ethernet statusBSS::AMBERMark Amber, CSC Network Mgr. (DTN)592-4645Thu Nov 07 1991 02:199
 >   evaluate.  All I am really trying to test is whether or not I can set
 >   up an alarm to tell me when a node goes down.  I want to use the
 >   ETHERNET_AM because it is faster than the DNA4_AM.  I plan to do the
    
    This might not give you the results you desire.  Remember that many
    [most?] ethernet controllers still send out SYSID's and respond
    to MOP REQID's even though the attached node may be "down".  Once
    it gets its physical address programmed in, many ethernet controllers
    will happily respond with that address as long as they have power.
1583.7Greedy for global Ethernet_AM station alarm ruleCUJO::HILLDan Hill-Net.Mgt.-Customer ResidentFri May 01 1992 05:4224
    The alarm rule mentioned in .3 should be modified thusly to work:
    
    EXPRESSION = (STATION .station.KANG -
    		RESPONDING ADDRESS <> AA-00-04-00-07-0C, AT EVERY 00:01:00), -
    
    This works well.  However, I'd like to create a GLOBAL alarm for
    determining reachability of certain nodes (like PCs and X-terminals).
    
    This expression works as a kludge:
    
    EXPRESSION = (STATION * BYTES RECEIVED < 0, AT EVERY 00:01:00)
    
    If the station entity responds, a value greater than zero will be
    returned and the alarm will not fire.  If the station does not respond,
    then an alarm of severity INDETERMINATE will fire.
    
    As I said, this is a real kludge, but it has allowed me to locate the
    odd "dead" X-terminal and LTM-bridge.
    
    Any ideas for doing this a better way so I can receive a CRITICAL
    severity?
    
    Thanks,
    Dan