cancel
Showing results for 
Search instead for 
Did you mean: 
Create Post
Level 10

Difference between Application Warning/Critical State & Application Down

Jump to solution

What is the difference between Application Warning/Critical State & Application Down?

I have never been fully clear on the difference.  What would an example of Application Down be?

Tags (2)
0 Kudos
1 Solution
Level 15

http://www.solarwinds.com/documentation/en/flarehelp/sam/default.htm#orionapmagstatusindicators.htm

Down = Can be reached but is not operational. (Polling request timed out.)

See bottom of page. HTH

View solution in original post

0 Kudos
7 Replies
Level 13

What is the difference between Application Warning/Critical State & Application Down?

Application Warning and Critical states manifest when a measured value passes a limit (threshold) that you've set. For example, if you have an Apache application monitoring an Apache server, you might set thresholds to notify you when your web server starts getting hammered.

You might set it up so any ServerRequestPerSecond value above 100 should merit a Warning, and any value about 150 should merit Critical.

Warning threshold: ServerRequestPerSecond >100 but less than <150

Critical threshold: ServerRequestPerSecond >150

What would an example of Application Down be?

Your Apache web server has crashed.

Level 15

Yes, when Warning & Critical thresholds are breached (can be user set) the status will reflect this. Also see the Orion Guide here:

http://www.solarwinds.com/documentation/en/flarehelp/orionplatform/default.htm#cshid=orionagstatusin...

Level 17

Usually a critical or warning are based on a statistical threshold being met where down is due to non operational as Bronx points out.

Level 15

http://www.solarwinds.com/documentation/en/flarehelp/sam/default.htm#orionapmagstatusindicators.htm

Down = Can be reached but is not operational. (Polling request timed out.)

See bottom of page. HTH

View solution in original post

0 Kudos

@aLTeReGo - Is this still the case in 2019.2+? I've got a 'Service Status - SNMP' component and the statistic threshold says 'Warning - not equal to null for a single poll' (since I don't really care about warning) and 'Critical - not equal to 1 for a single poll'. When the value comes back not equal to 1, the status of the application is 'Down', not critical.

Is this just valuable for components where thresholds might be distinguished between warning and critical (such as file count monitor)? If so, do I even need to put in a value for either of these or just define the 'Net Service Display Name' and the interval (i.e. for a single poll OR for x out of y polls) and save it like that. Then expect that the application will always go 'Down' if a service component is 'not equal to 1' per se?

Should this article be updated to reflect this behavior as well?

http://www.solarwinds.com/documentation/en/flarehelp/sam/content/sam-service-status-snmp-sw3296.htm?...

0 Kudos

@bhumphrey wrote:

@aLTeReGo - Is this still the case in 2019.2+? I've got a 'Service Status - SNMP' component and the statistic threshold says 'Warning - not equal to null for a single poll' (since I don't really care about warning) and 'Critical - not equal to 1 for a single poll'. When the value comes back not equal to 1, the status of the application is 'Down', not critical.

Is this just valuable for components where thresholds might be distinguished between warning and critical (such as file count monitor)? If so, do I even need to put in a value for either of these or just define the 'Net Service Display Name' and the interval (i.e. for a single poll OR for x out of y polls) and save it like that. Then expect that the application will always go 'Down' if a service component is 'not equal to 1' per se?

Should this article be updated to reflect this behavior as well?

http://www.solarwinds.com/documentation/en/flarehelp/sam/content/sam-service-status-snmp-sw3296.htm?...


The application status should only be 'down' if the service isn't running at all. As configured, your application would only be in a 'Critical' state if the service was in one of the following states. 

2 - Continue Pending;
3 - Pause Pending;
4 - Paused.

Also, if you don't define a value for a threshold, then it's ignored. E.G. the following would be how I would configure the thresholds. 

 

 

 

 

Ah, that makes sense. I was assuming that all of the values that could be returned were on that list. Thank you for the clarification!

0 Kudos