This discussion has been locked. The information referenced herein may be inaccurate due to age, software updates, or external references.
You can no longer post new replies to this discussion. If you have a similar question you can start a new discussion in this forum.

AppInsight For SQL

Hello Everyone,

I recently started using solar winds, so far i think this tool is awesome. But like with every tool, i have questions..

I've installed SAM 6.1.0 and have used the network sonar discovery to find my SQL cluster nodes and the named instance vips. I then chose the AppInsight for SQL resource to monitor the nodes and vips. I've let the machines poll for a while and have noticed the following.

Each of the nodes i've added end with SQN each of the instances end with SQD. I was expecting to only see information on the instances that are currently being hosted on the node in the picture below, but i see every instance on the cluster. Does anyone else see this in their implementation? Should i just ignore the greyed out instances? Is there a way to only show the active instances on the specific node?

Solarwinds1.png

On another note i also found that the AppInsight for SQL template does not monitor disk size for the SQL nodes. This is alright i ended up creating a custom template to monitor system, non system and mount point drives, that should work for all servers in my environment.

Thank you for your time.

  • Hello Everyone,

    I ended up finding out that the greyed out instances were just not setup correctly. They were using the default port instead of the actual port that all of my instances are using. After entering in the right information the grey instances turned red with a note beside them saying "never up". At this point im stuck on how i can tell what instances are active on the specific node im looking at. Is there a switch somewhere to tell AppInsight for SQL to only report on the active instances of each node but have each node know about each of the instances that could potentially move to it?

    Thanks,

  • Based on what you've described and the screenshot provided it looks like maybe you tried to assign AppInsight for SQL to a cluster member, and not the cluster VIP. Is that what's going on here?

  • Hello aLTeReGo,

    What I did was the following.

    1. Clicked Network tab, Manage Nodes, add Node

    2. Completed a discovery, looking for all my VIPs for the SQL instances

    3. It returned my 4 cluster nodes SQN01 - 04
    4. I added the resources you see below to each of the SQN (SQL Cluster Nodes) servers.


    ListResources.JPG

    I have a feeling what i should have done was looked for each VIP and added AppInsight to the VIP instead of the SQN nodes?

    Thank you,

  • That is correct. AppInsight for SQL should not be applied to the individual cluster members. It should be assigned only to the shared VIP. AppInsight for SQL will give information related to the cluster and its members, but should not be assigned to each cluster member individually. By not requiring you to assign AppInsight for SQL to each cluster member individually you also save a bunch of component monitors. emoticons_happy.png

  • Strange, because when i tried to add the VIPs as part of the network discovery, it only came back with the four nodes. I may be able to add the AppInsight for SQL template to the cluster VIP which would help do the same thing. I'll let you know what it comes back with.

  • Instead of using Network Sonar, try simply adding the node individually and using "List Resources" to select AppInsight for SQL.

  • You my friend are a genius!

    I'll write up a document and post it on here with the exact steps on how to setup clustered instances. Because i've moved from using system centre operations manager to using solar winds, i was used to the concept of being able to monitor servers, and instances, and disk space all in one bang. But with clustered instances in solar winds with this specific AppInsight for SQL template you should only focus on monitoring the instance. Other templates can be used to monitor disk space and server side issues. I've created my own custom template for monitoring system, non system, and mount point disks via WMI. So now i think I'm good to go.

    Thank you very much alTeReGo emoticons_happy.png

  • Love to see what you write up. If it qualifies, I'll put it in the next version of the Administrator Guide, if it's not already there.

    While we're on the subject, if you (or your SolarWinds friends) have any other tips you'd like to share, please post them here: Comments Needed! Always looking to add quality information to the Administrator Guide for SAM.

  •   

    We have got cluster SQL servers in SQL Alwayson... for example, instance X is available on 3 SQL servers but active only on 1 server. So in this case how do i need to make use of Appinsight for SQL?

  • When Configuring AppInsight for SQL you would point it at the name of the listener for the AlwaysOn cluster, no differently than you would an application that was using a database on the AlwaysOn cluster.