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

An alert that won't alert

Hi:

It’s been several days that I’ve been knocking my head against the wall… I just can’t figure this one out. I have an alert that works when I test it but in real life it does not trigger.

The monitor watches a folder and goes into critical when a file older the 5 minutes exists. So what I do is:

1. Place a bad file in the folder.

2. Wait the 300 or so seconds…

3. See the component and the application go into critical…

4. Wait and wait and wait… the alert does not trip.

My first thought was I have my trigger messed up but since it works in the test and when I grab the exact query from the alert definition it shows a hit... I'm thinking it something else.

The triggered actions work because when I test they work.

My application/monitor stays critical and no action takes place. (BTW… other actions/alerts are running just fine and yes this alert is enabled). Any ideas on what I could be the problem?

orion.jpg

Find more posts tagged with

Accepted answers

gbutler

I need to correct what I stated above. The reason the email was not being sent was NOT because the action errored. It was because I had the option "Do not execute this action if the alert has been Acknowledged" checked. Now I did not acknowledge the alert so I don't know why it would make a difference... but now that I have it UNCHECKED. I can run with two defined actions. I tested and if action ONE dies action two still runs. Yah just can't check that box!

AAAA.jpg

All comments

gbutler

I opened up a support ticket so hopefully they will point out my blunder. I did find some pretty cool things by watching the SQL stuff in the database. I got tired of watching/waiting for that stupid email that would never arrive so I set up some standard queries that would watch - the alert log, my alert's status, alert definitions. Anyway for those who are newer then I am I thought I'd share:

orion.jpg

gbutler

Bingo. I think I have it. Per my DB query the alert was actually triggering. So the problem is not the trigger but the action. I have multiple actions. The first action would fail... so the second action would not run. The only way it would run is if you reset the alert (I would do this by adding/deleting a new condition into the trigger... this would warn me that the alert will be cleared). Once it was cleared actions would take place.

Lesson to learn is be very careful when you have multiple actions. When my vbscript would error on a non-zero value all goes to a stand still. Also, make sure your email alert preceeds any script you have in the alert action list.

I'm not sure I like this "feature". If Orion is capturing the ERROR wouldn't it make sense to log the error somewhere or report it? Well written scripts will sometimes error and under these conditions no one would be the wiser until something really breaks. IN summary, you can have an alert that:

1. Passes a test when created giving the user confidense that things are good.

2. Fails in production when stuff really breaks.

I'm just lucky that I wanted to test it manually and not rely on the tests provided.

gbutler

AAAA.jpg

vinman1000

I had a similar problem with SNMP and syslog alerts not performing their action. Once I restarted the service (SNMP or syslog respectively) it works. Every time I make a change I have to restart the service to get the alert to work.

Worth a try...