How about this: catch the threshold in a rulelset, and don't send it if it
rearms within so many minutes?

Cordially,

Leslie A. Clark
IBM Global Services - Systems Mgmt & Networking
Detroit


                                                                                
    
                    netview@toddh.                                              
    
                    net (Todd H.)        To:     IBM NetView Discussion         
    
                    Sent by:              <nv-l AT tkg DOT com>                 
           
                    owner-nv-l@tkg       cc:                                    
    
                    .com                 Subject:     [NV-L] Data               
    
                                          collection/thresholds, ruleset 
question,  
                                          trap content                          
    
                    09/19/01 09:07                                              
    
                    PM                                                          
    
                    Please respond                                              
    
                    to IBM NetView                                              
    
                    Discussion                                                  
    
                                                                                
    
                                                                                
    




I have been asked to monitor CPU% on a router and send an alert if the
% exceeds 80% in "two consecutive polling intervals-5 minute
intervals".

Now, given that the Thresholds tool is edge-triggered (i.e. goes off
once, and doesn't reset until the rearm level is satisfied), is there
any way to implement this request?

At first, I though the "Threshold" node in my ruleset with an AT count
of 2 might do it.  However, then I realized that the data collection
tool would only throw the IBM_NVDCOL_EV trap once until the rearm
level was satisfied.

Is there a workaround, or do I have to go back and sell the requestor
on the benefit of seeing an alert immediately upon exceeding a CPU
threshold?

Best Regards,
--
Todd H.
http://www.toddh.net/



_________________________________________________________________________
NV-L List information and Archives: http://www.tkg.com/nv-l