Conference azur::mcc

4495.0. "Alarm rule help" by CGOOA::KUHNEN () Wed Feb 03 1993 14:28

    Hello,
    
    
    I am running DECmcc in a network in which there are no redundant paths.
    I am struggling with how to recognize that there has been a failure in
    the path to remote stations that I am monitoring, so that the alarm
    rules monitoring these remote stations do not cause the mangement
    station to choke with the failures when the path to the remote entities
    has disappeared. All rules fire with exception paramters when this
    occurs.
    
    Has anyone written command procedures which disable rules when a major
    router goes downs , checks for reachability and then reenables the
    alarm rules when the path is reenabled. If so would you mind posting
    the code so that I could use it as a template.
    
    I would also like to comment on the occurs n time rules. Since the 
    Ipreachability tests are not very reliable I must use one of these
    rules to check for reachability. This rule however fires more often
    than I would like due to the fact that a clear notification is counted
    as an OSI rule firing. 
    
    For example I check reachability on the node with a repeat interval of
    1 minute. I then set up a rule that will fire if the first rule fails
    4 times in 4 minutes. My rule will fire if the node is first
    unreachable, then reachable , then unreachable and then reachable
    again. Every time there is a change back to reachable the rule fires
    and this is included in the counters that checked in the second rule.
    Is there a way to get around this situation.
    
    Thanks
    
    Harry Kuhnen
           

    I am working on a set to do that utilizing the alarm managers
    FCL batch job so that I can enable and disable rules
    from the command procedures.  These aren't done but if
    you want more info on what I am doing give me a call.
    
    jil