KS-Soft. Network Management Solutions
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister    ProfileProfile    Log inLog in 

TestA starts TestB -- daisy chaining tests

 
Post new topic   Reply to topic    KS-Soft Forum Index -> Wish list
View previous topic :: View next topic  
Author Message
avery_larry



Joined: 11 Sep 2008
Posts: 25

PostPosted: Thu Jul 10, 2014 4:02 pm    Post subject: TestA starts TestB -- daisy chaining tests Reply with quote

There are some ways to mostly accomplish this, but as far as I've been able to find, this doesn't quite exist. I'm happy to be wrong if there's a great way to do this . . .

What I'm interested in is a master test which will start a dependent test immediately after each successful test (or failed test, or whatever).

The best example (and the main reason I'm posting) is a traceroute style chain of tests. If I have a chain of 15 hops between myself and a remote location, I'd like to ping hop 1, and if it's up then immediately ping hop 2, and if it's up continue down the line. Then I can tell which hop fails with a MUCH smaller window of error. In other words, I will often get hop 10 to test as down when it's actually hop 6 that's down, but because of the timing of when the tests fire, I'll still get a down email for hop 10, hop 14, and hop 6. For that matter, sometimes hop 6 won't even give me a down alert because it'll be a brief outage that's rectified before the scheduled interval for hop 6.

Current options:
1) Set increasing intervals for each hop.
Problem: It decreases the issue somewhat, but not very much.
Problem: The larger the increase in interval, the more effective this approach is, but after maybe 3 or 4 tests in the chain the scheduled interval gets to be larger than desired.

2) set some combination of the "Behavior" settings "Recheck dependent ..." and "Consider status ... obsolete after __ seconds".
Problem: The huge main problem I have with this is that these are *global* settings. I have (as I'm sure most people have) hundreds of tests of all varieties that depend on master tests. I don't consider these settings to be appropriate for all the different variety of tests. For my disk space checks, the obsolete status can be a minute (or longer). For the traceroute concept, it should be 5 seconds (maybe less). Setting it low globally will cause extra unnecessary tests.
Problem: This setting works a bit backwards from what I'm asking for. The last test in the chain will basically work it's way to the beginning of the chain forcing them all to be current.
Problem: If I try to chain a lot of tests together, I *think* I could run into a problem where the obsolete setting is set so short that I can never fit all of the tests into the time window.

3) Reverse the master/dependent order. I could test the last hop, and then set the hop before it to be dependent and trigger when the master test is dead or unknown, and work my way forward.
Problem: Getting the alert profiles to work correctly is -- at best -- counterintuitive.
Problem: Working with multiple traceroute chains, The first hop is common to all, the 2nd hop is common to most, the 3rd hop is common to some etc. I'd have to setup multiple tests for the same hop -- one for each chain, which will result in multiple alerts.
Problem: Well, it's really philosophically backwards in the "Master, dependent" relationship concept.


For my money, I'd like this to be added into the master/dependent test concept. You have a couple of synchronize options already setup (counters/status/alerts), just add a "synchronize schedule" option where the dependent test's schedule immediately follows the master test's completion (and matching the condition).

Another place you could add it, instead of the master/dependent test concept, is the action profiles. The alive or dead action could be "perform test ______".
Back to top
View user's profile Send private message
KS-Soft Europe



Joined: 16 May 2006
Posts: 2832

PostPosted: Thu Jul 10, 2014 4:21 pm    Post subject: Reply with quote

Have you tried HMS Script action with commands like:
Quote:
RefreshTest <TestName> Checks the status of the specified test immediately
RefreshTestByID <TestID> Checks the status of the specified test immediately
RefreshIrregularTestByID <TestID> Command works similar to RefreshTestByID command - it tells HostMonitor to perform test probe immediately. In contrast to RefreshTestByID command, it can force execution even for test items that were configured with Irregular schedule

Please check for details at:
http://www.ks-soft.net/hostmon.eng/mframe.htm#actions.htm#actScript
Back to top
View user's profile Send private message Send e-mail Visit poster's website
avery_larry



Joined: 11 Sep 2008
Posts: 25

PostPosted: Mon Jul 14, 2014 3:46 pm    Post subject: Reply with quote

I will check it out. It does sound like it would do what I'm asking.
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic   Reply to topic    KS-Soft Forum Index -> Wish list All times are GMT - 6 Hours
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group

KS-Soft Forum Index