If the product is so dependant on it, why isn't there a service to check for a SQL down error to the database?
Solarwinds Services log some events in the Windows Application Log Event when a such error occurs.
Are you running a 2nd monitoring tool to watch the SQL server that ORION is dependant upon?
This would definitely be a good and relevant solution.
A small script running on a third server that checks periodically the status the SQL Service and sends an email in case it is stopped or in an unknown status might be sufficient. However relying on a solid 2nd monitoring solution would make you more confident.
The problem remains the same with a local SQL Express database, if the service stops, your alerts will not be triggered as the AlertingEngine will be in faulted state. I do not think having the SQL Server locally or remotely changes something to the issue.
So why don't you (Orion Solarwinds) create a service to do this? I see this as a short coming in your application.
You are telling me I need to buy a second product to monitor your product?
The problem remains the same with a local SQL Express database, if the service stops, your alerts will not be triggered as the AlertingEngine will be in faulted state.
I have noticed that is the SQL is not reachable: system down, network path broken, etc... Then avery alert I have triggers and I get bombarded with hundreds of false alarms until the SQL becomes available. I have seen other posts on this as well.
I would love it if the alert engin would go into a faulted state. How can I fix this?
We use Nagios to monitor our SolarWinds systems. IIRC, SolarWinds APM has a built-in thing for monitoring SQL Servers, so you could use that. But since our SolarWinds is considered very important, we wouldn't want to risk expecting a partially functioning SolarWinds to notify when part of itself goes down.
For the actual nagios check script for the database, I use a little python script which just connects to the SQL Server, runs a query, and reports if the database is alive. Not somplicated at all, but it seems to be working great. We also have the Nagios nsclient++ agent installed on the main Solarwinds machine to monitor the state of all the solarwinds services.