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
Store

Including Pertinent Information in an OpsGenie Alert

I am relatively new to SolarWinds, having come from a deep background in System Center Operations Manager. I seem to be struggling with the way SolarWinds handles alerts.

I am trying to build a general alert to notify when a component goes "Critical". I have built the alert and have it working fine with OpsGenie, sending each variable I need over to the integration - almost.

I want the description of the alert to contain the information about what is wrong. One complaint about the environment I have taken over is that alerts didn't have relevant information; you still had to look somewhere else (like SolarWinds to see how bad something was. For example, in the case of a Logical Disk: Avg. Disk Queue Length critical event, I would like to see:

The metric Logical Disk: Avg. Disk Queue Length on XXX-P-XX-XXXX is at a value of 11.2580280303955 Avg. Disk Queue Length

I am passing the following variables to OpsGenie (amongst others):

Entity=${N=SwisEntity;M=Application.Node.Caption}&

ComponentName=${N=SwisEntity;M=ComponentAlert.ComponentName}&

ComponentNumeric=${N=SwisEntity;M=ComponentAlert.StatisticData}&

ComponentValue=${N=SwisEntity;M=Application.ApplicationAlert.ComponentsWithProblems}

When I create the description using the variables, it looks like this from a code standpoint (in OpsGenie):
The metric {{ComponentName}} on {{Entity}} is at a value of {{ComponentNumeric}} {{ComponentValue}}.

For systems where I have one faulting component, this works fine. However, I noticed that I had the below output on some of my alerts:

The metric Logical Disk: Avg. Disk Queue Length on XXX-P-XX-XXXX is at a value of 11.2580280303955 Pages/sec(Critical), Logical Disk: Avg. Disk Queue Length(Critical), Physical Disk: Avg. Disk Queue Length(Critical).

Does anyone have any insight into what I may be doing wrong? Is there a way to repeat the description above and grab each Name, Numeric, and Value?

Find more posts tagged with

Accepted answers

mesverrum

The problem in this case is that you are using the variable for components with problems inside an alert that should be for a single component. So it shows you the component that triggered the alert, as well as any other components that have issues on that application at that time. Unless you are using a complex alert condition you shouldn't need the braces <<<>>>. What data are you actually trying to display as your "component value" because it looks to me like you should just cut out that entire variable, unless I'm missing something. Seems like you already have the info you need in the first few variables.

All comments

rfroembgen

Hi,

i got some great input from mesverrum which could also help in your case. He pointed me to a "hidden" function Orion does if you wrap variables in braces like <<< ${N=SwisEntity;M=Application.Node.Caption} >>>. It makes Orion return whatever is available for this variable based on your trigger condition.

There is a default alert (polling rate limit exceeded) which you could use as an example.

Best Regards

Rene

mesverrum

sghohman

Thank you for the replies.

I am trying to drop in a variable to tells the recipient what the metric is that they are seeing. For example, if we alert on low disk space and a value is sent, I want the recipient to know that the value is % Free Desk Space, or Free MB by looking at the email.

I guess the Component Name variable would probably viable. I think I need to test more...

Still learning - thanks again for getting the gears turning.

mesverrum

Yeah, my typical layout for a component alert is usually something like this:

Component ${N=SwisEntity;M=ComponentAlert.ComponentName} is ${N=SwisEntity;M=Status;F=Status} on ${N=SwisEntity;M=Application.Node.Caption} - ${N=SwisEntity;M=Application.ApplicationAlert.ApplicationName}

Current Statistic Value: ${N=SwisEntity;M=ComponentAlert.StatisticData}

Critical Statistic Threshold: ${N=SwisEntity;M=ComponentAlertThresholds.ThresholdStatisticCritical}