Comprehensive IP SLA scheduling capabilities and reaction trigger capabilities

Question

There are many key topics I would like to discuss as features that we currently use with other IP SLA monitoring tools and very much like to transition these permanently to Solarwinds.

Item 1

Scheduling capability of IP SLA is fairly rich featured but current methodology available when using your SNMP SET capabilities is standard, start now, life, forever.

Item 2

Using the TAG field is common and often required to allow scheduled IP SLA pats to be used across more than one NMS system and additionally, makes it much easier to annotate in the setup of the router information about the type of test.  Additionally, it is very important that solarwinds sets the OWNER field on the router when setting up these tests as well as support for showing that field when learning existing tests.

Item 3

Reaction trigger is a very useful feature for enabling traps such that ORION can get a more real time alert when the test passes some limit..common example is Packet loss exceeding 10% on UDP VOIP Jitter test.

FormerMember · Answer

Thanks for your reply.  The most common use of scheduling is related to hours of operation for a site.  It isn't usually necessary for locations that are essentially closed/unmonitored to continue to perform IP SLA operations.  So an example would be that you would set the schedule to be Mon - Fri 7 AM to 7 PM Pacific...as an example.  This is a well documented method by Emanuel Tychon and taught in the IP SLA Advanced class.

Item 2

TAG is required when your IP SLA test is also shared by your provider.  Example, Sprint requires that OWNER is setup to reflect the appropriate name of the test so that their system, VitalSuite can monitor the Test.  Why setup 2 tests.  Since many people are establishing tests to include those manually entered via Command line and those entered by Solarwinds through RW strings Owner and Tag are necessary.   More on tag:  We use Tag to describe in plain english the test so that Alerts and Alarms are more meaningful to the NOC.  Example would be TAG:  UDP Jitter test from Cleveland Clinic to Akron Clinic...instead of 10.0.0.1 UDP Jitter 10.0.0.2...uggh that's just awful...remember people use the command line on the router too..and the output devoid of owner makes using the ip sla monitor commands unbearable.

Item 3:  I think you are confusing the difference between As polled versus Syslog/Trap.  Reaction trigger is syslog or Trap related but it's a part of IP SLA code.  So As polled might be every 5 minutes..but Reaction Trigger would be as immediately seen by router....very different use case..but all IP SLA.  Thus a TRAP sent using Reaction Trigger would be good to correlate to statistical aspects of collected data from SAA MIB.

Hope this clarifies.  I would like to talk to you about this in much greater detail so let's setup some time soon.

macnugetz · Answer

Item 1

What would you like to see in terms of scheduling capabilities?  In other words, what's your use case around schedule operations?

Item 2

Could you give me a specific example of how you would use TAGs?  Also, why is it important to set the OWNER field for yoU?

Item 3

You should be able to configure alerts like this in IPSLAM today.

-Craig