cancel
Showing results for 
Search instead for 
Did you mean: 
Create Post
Level 16

Can we alert for SNMP not responding nodes?

Jump to solution

Hi all,

I searched about this topic but all are very old so wanna know if we can alert for nodes which are not responding to SNMP.

I have only 5-6 nodes for which alerting is required.

So can anyone help me on this?

We actually had an issue for some devices where they were found in hung state and they had reboot the device to bring to normal. So according to me setting up non responsiveness to SNMP will atleast meet this requirement.

Tags (1)
0 Kudos
1 Solution
Level 8

This alert seems to be working fine for me. This will send an alert for any device that is not "Down" when Hardware or child node status is "Unknown" for any polling method. If SNMP stops responding, the child status and/or hardware status should change to unknown and trigger the alert.

pastedImage_0.pngpastedImage_1.png

View solution in original post

0 Kudos
28 Replies

You may find the simplest, most straight-forward solution is to put the Hardware Health Overview window on your front page.  It'll show any devices that stop responding to snmp as "Undefined", and will update automatically in real time.

pastedImage_0.png

To get this into your home page, open Orion NPM. The top right corner of the page should have a link called "Customize Page". Click on it and add the "Hardware Health Overview" to any of your columns, and move it up or down to the position you think makes best sense.

pastedImage_0.png

Then all you need do is click Done, and watch for that green circle to show a Gray line, and for the gray "Undefined" count to show anything greater than zero. Click on the "Undefined" link to see which nodes have stopped responding to snmp.

I like this idea (hardware status unknown), but it does not seem to work for Linux virtual appliances. Would it be accurate to say that these Linux vm's do not actually have a "hardware status"? That's what it looks like to me. If this is correct, then is our only option to do something like this with LastSystemuptimepollUTC:

WHERE (s.LastSystemuptimepollUTC < DATEADD(HOUR,-72,GETUTCDATE())
OR s.LastSystemuptimepollUTC IS NULL)

 

Thanks in advance!!

0 Kudos
Level 8

This alert seems to be working fine for me. This will send an alert for any device that is not "Down" when Hardware or child node status is "Unknown" for any polling method. If SNMP stops responding, the child status and/or hardware status should change to unknown and trigger the alert.

pastedImage_0.pngpastedImage_1.png

View solution in original post

0 Kudos

This worked perfectly for me.... was able to test it yesterday for a new customer... thanks for sharing it,,,

0 Kudos

Can you show us configurations image ?

Enviado do meu iPhone

Em 10 de nov de 2016, às 17:49, jtranchina@mathematica-mpr.com <communityteam@solarwinds.com> escreveu:

Can we alert for SNMP not responding nodes?

reply from jtranchina@mathematica-mpr.com in Network Performance Monitor - View the full discussion

This alert seems to be working fine for me. This will send an alert for any device that is not "Down" when Hardware or child node status is "Unknown" for any polling method. If SNMP stops responding, the child status and/or hardware status should change to unknown and trigger the alert.

https://thwack.solarwinds.com/servlet/JiveServlet/downloadImage/2-349466-136966/pastedImage_1.png

Reply to this message by replying to this email, or go to the message on THWACK

Start a new discussion in Network Performance Monitor by email or at THWACK

Following Network Performance Monitor in these streams: Email Watches, Inbox, Connections Stream

0 Kudos

It's in the post....you may be having issues viewing it on your iPhone.

0 Kudos

Ok I see now If you set status different up? Ane test a single node after for more .

Enviado do meu iPhone

Em 10 de nov de 2016, às 18:06, jtranchina@mathematica-mpr.com <communityteam@solarwinds.com> escreveu:

Can we alert for SNMP not responding nodes?

reply from jtranchina@mathematica-mpr.com in Network Performance Monitor - View the full discussion

It's in the post....you may be having issues viewing it on your iPhone.

Reply to this message by replying to this email, or go to the message on THWACK

Start a new discussion in Network Performance Monitor by email or at THWACK

Following Network Performance Monitor in these streams: Email Watches, Inbox, Connections Stream

Following Can we alert for SNMP not responding nodes? in these streams: Inbox

0 Kudos
Level 13

Hello,

you can use for some critical nodes SNMP status polling.

List resources on that nodes and instead of status polling via ICMP check SNMP status polling.

LAda

0 Kudos

yes this option i m aware but dont want to use it.

0 Kudos
Level 16

Will this work, can you help me confirming?

Capture.PNG

0 Kudos

Good day, you may wish to consider "Unknown" and Polling Method as shown below:

pastedImage_0.png

one thing though..if i use the same setting but for ICMP then i think that will also alert me if the node has gone into unknown state.

Correct me if m wrong.

0 Kudos

It should work fine; of course Simulate against a node and get results before actual save.

Hi smoked_angus

But I don't see any option of 'unknown' for Polling Method as shown by you, are you sure its the same? OR were you referring to Node status?

0 Kudos

Yes, it is there, with polling method.

0 Kudos

Oh..k let me check..and alerting Wil be on Node type right?

On 05-May-2016 5:33 pm, "smoked_angus" <

0 Kudos

Cool...

0 Kudos

I think this should work for me...didn't know that stat Collection and polling method variables are also available. Thanks a lot 🙂

I will try this and let you know....

0 Kudos

You can also just force these nodes to poll solely through SNMP, if that would work as well. Then you'd get a down alert when you don't get an SNMP response.    

0 Kudos
Level 16

I use a custom SQL alert on nodes.

Where

[status] in ('1' , '3' )

and unmanaged=0

and DATEDIFF(MINUTE,LastSystemUpTimePollUtc,GETDATE()) >60

so if the think stops responding to SNMP polls of it's lastSystemUptime for an hour then we get an alert for it