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

Node is in a Warning or Critical state

On Orion version 2020.02.1276

Have windows servers monitored with Windows agent. One of the servers is a hyper-v host for the other guest. Everything was showing green with no alarm triggered.

On the hyper-v host I went into the "List Resources" I enabled the "Hardware Health Sensors" and the "Hyper-V" choices.

After that I think possibly (not sure) all of the guest servers are now showing a critical status and I can't figure out why.

For example a guest server will show node status: Node is Warning. If I click on the alert it show current status: triggered. Severity: Critical.

Nothing I can see shows what it is complaining about.

If I clear the alert I comes back.

On the hyper-v host I removed the "Hardware Health Sensors" and the "Hyper-V" choices in case that was the cause of it. However I still can get the alerts on the guest to clear.

Currently the hyper-v host itself is showing healthy with no alerts.

Find more posts tagged with

Accepted answers

aLTeReGo

Well I think we've successfully ruled out all the easy things. At this point I would recommend opening a case with support so they can delve deeper into the issue. If you'd like to continue troubleshooting on your own though, there are two other things you can try.

#1 Verify you have no alert actions using 'Set Custom Status'. This is one way where you can force the status of a node to be anything you like based upon the trigger condition.

#2 Try disabling each Child Status Contributor one by one to see if you can determine which child entity type is causing the node to be in a warning state.

Before doing this, I recommend disabling alert actions just in case. This is just a precautionary step and should be re-enabled after you're done testing.

Once alert actions are paused, go to [Settings > All Settings > Node Child Status Participation]

take note of which Node Status Contributors are enabled/disabled before going further so you can revert these changes once you identify entity type is causing the node to be in a warning state. I usually take a screenshot as the fastest/easiest way of documenting these settings before making changes.

Disable each node status contributor individually, waiting a minute or two between each and checking to see if the status of the node changes from 'warning' to 'up'.

All comments

aLTeReGo

Are you monitoring the guest VMs as nodes in Orion? If so, what do you see when you hover your mouse over one of the guest VMs that's in a critical state? If possible, please provide a screenshot and obfuscate any sensitive information like the machine name or IP address.

wseyller

Yes the guest VMs are monitored as nodes and each have the Orion agent software.

There is one host server with a total of 5 guest VMs. The host and the guest are monitored separately as nodes each with their own Orion agent installed.

The host and two of the VMs (one of the domain controllers and the orion server) are not showing any alerts.

Three of the guest are all showing the same alert.

aLTeReGo

What version of the Orion Platform are you currently running? Also, do you have Enhanced Status enabled?

wseyller

Looks like the Orion version: 2020.2.5

The node status is set to enhanced.

adam.beedell

I think alterego guessed the same thing i'd guess first - The VM has a poor status, which is linked to the node layer, but is otherwise distinct and doesnt show up easily when the node layer object is asked. VM poor statuses when there isnt an obvious cause is probably due to an alarm in VMware/HyperV. You can verify this by searching (top right search) for the caption, if there's two results back that both link to the same thing you've got VM-level information. This can be verified in the DB too.

If it's not that (and i swear it's usually that) it should show on hovering over the object, if it doesnt then look for objects which are linked to it, like a VM (an application for example would cause this in the same way, but is much easier to tell in the UI.

In Enhanced node status it should show the contributing objects, one in there that also isnt represented super well i the UI as default is API pollers. If so you can add a resource to the page/check Manage API pollers, or under the management widget there's a "manage api pollers" option. I dont think this is the solution though.

I'd put a fiver on an alarm from virtualization layer onto VM object as adjacent table to the actual node which you're checking

wseyller

I went into the node status contributors and I turned off "Virtual Machine". So I am going to leave that off. Thanks

aLTeReGo

Sounds like that resolved the issue?

wseyller

Yes it is still working good now. So yes it is resolved. Thank you.