Community
- Command Central
- MVP Program
- Monthly Mission
- Blogs
- Groups
- Events
- Media Vault
Products
- Observability
- Network Management
- Application Management
- IT Security
- IT Service Management
- System Management
- Database Management
Content Exchange
- SolarWinds Platform
- Server & Application Monitor
- Database Performance Analyzer
- Server Configuration Monitor
- Network Performance Monitor
- Network Configuration Manager
- SQL Sentry
- Web Help Desk
Free Tools & Trials

Application Monitor

We've bought Application Monitor and I was wondering if somebody else out there was experiencing the same as me ragrding the number of events.

I estimate that the number of events is now about 10 times as high as before we started using Application Monitor. The vast majority of the events are "application stopped" or "application restart"

Lars Grabowski
Global Infrastructure
Chr. Hansen A/S

Currently using:
Cirrus Version 2.1 DL-500 Feb. 2006 (8.1.42)
Orion Application Monitor 7.8.82
Orion Web Engine Version 7.8.92
Orion NPM SLX 7.8.2 July 2005
Alert Manager 8.1.15

Find more posts tagged with

Accepted answers

All comments

dpezet

I have to say, if there is one thing I am dissapointed about in Orion it is definitely the Application Monitor pack. The lack of integration with Orion is impressive.

That being said, all the alerts I have received from it have been true. When it says an app is closed, the app is definitely closed on the server. If you are receiving excessive or eroneous alerts you may actually be having SNMP issues and not issues with the Application Monitor. I had a handful of Windows 2003 servers (maybe 3 out of 60) that would not work with the Application Monitor until I removed SNMP and reinstalled it. Everything else worked fine. It seemed to be a problem with the Host Resources MIB on those servers.

There is another thing to consider. I have an application that crashes periodically for no reason. We are working with the vendor on that, but in the mean time I have the service set in windows to restart if it crashes. I just used the normal setting in the "Services" administrative tool under the "Recovery" tab for the service. Now, if the program crashes Windows will restart it and the Application Monitor will detect that it restarted and alert me. I assume it does that because the PID has changed, which in my opinion is a good thing. So, if the PID changes on your app then that may cause the alerts you are seeing.

You may also want to consider setting up supression on the alerts. If a server goes offline you do not want to receive one alert for the server going down and then multiple alerts for the individual apps on the server going offline. I setup supression on the application alert to only alert me if the "NodeStatus" was equal to "Up" for the server the app is on. I just assume if the Node is having problems then the apps are too.

Hope that helps,

Don

Grabowski

Thanks Don,

Have you worked out what the difference is between "Applicatio xxx started" and "Application xxxx REstarted"?

I have set up suppression and it does work for me, but it "only" reduces the number of alerts triggered. It does not reduce the number of events generated

When I look at my application events, it seems that the vast majority of these are coming from servers in (very) remote locations (half way across the world typically) on links that are not among our best. I am going to experiment with the trigger and reset delays in the Beta Alerting Engine , which most likely will quiet things down for our on-call team, but I don't expect that the events-lists in Orion will become less crowded because of this.

How often do you poll you applications?

It would be nice if some sort of suppresion of the events could be handled by the Beta Alerting Engine, or if the same kind of agent that solarwinds has in the SysLog server (deleting events that meet certain criteria) could be added.

Lars Grabowski
Global Infrastructure
Chr. Hansen A/S

Currently using:
Cirrus Version 2.1 DL-500 Feb. 2006 (8.1.42)
Orion Application Monitor 7.8.82
Orion Web Engine Version 7.8.92
Orion NPM SLX 7.8.2 July 2005
Alert Manager 8.1.15

lavelanet

I have to agree that was the one thing that killed the app pack for me was all the events generated (pushing important events of the top 25 list) when oracle spins up and shuts down processes on unix boxes. I don't need to know about all the times oracle is starting or stoping a process. I wished it had finer control over what you needed to know.

Grabowski

I found a way around the host of application events. Not a brilliant one, but at least it works for now.

I created two reports in Report Maker. One that only includes events where the "Event Type Name" does not contain "application" and one that does NOT contain "application", Then I customized the view on the website to show these two custom reports. The only thing that is missing is the colouring of the backgrounds relating to what Event Type it is.

One thing that REALLY puzzles me though, is that when you create a report for showing Events, it is not possible to filter is by custom properties. I have seen that on a lot of reports that

Any help here would be appreciated, especially because it seems that the account limitations have no effect when a custom report is run, therefore I get ALL nodes shown on accounts where limitations are working normally.

Lars Grabowski
Global Infrastructure
Chr. Hansen A/S

Currently using:
Cirrus Version 2.1 DL-500 Feb. 2006 (8.1.42)
Orion Application Monitor 7.8.94
Orion Web Engine Version 7.8.133
Orion NPM SLX 7.8.5 March 2006
Alert Manager 8.1.15