Reboot Alert - Why so silent?

NPM 12 upgrade reboot server alert is not working - SolarWinds Worldwide, LLC. Help and Support

It turns out, I just needed to reboot my server, but here's the steps I executed against my server.

1. Run a repair on Core Services in control panel.

2. After the Configuration Wizard runs, stop all services.

3. Open Database Manager.

Run queries against the database:

delete from dbo.PendingNotifications

delete from dbo.SubscriptionTags

delete from dbo.Subscriptions

4.Reboot the server.

The funny thing is, I repaired the core service due to an assembly missing in the windows directory, which lead me down this path to begin with.

You can also follow the below link, same info is given there.

I'll test it out next month when I need to reboot it under normal circumstances and see if the changes this made actually fixed it. This is probably the third go around with me clearing the tables and rebooting, but never repaired the core services until now.

-R

All comments

We have the same anomalies.

How clean are your Windows App/System logs for SolarWinds related events?

The 'Solarwinds.Net' application log is pretty clean, no warning or error logs in that one. The system log has a few DistributedCOM logs for things that have been previously deleted from the node inventory, though I feel it's unrelated for the most part. I'll get around to figuring out why those are in there later. I had some netpath errors in the application log, but was mostly due to an SSL handshake error with the URL I was monitoring. I've since removed it and those have stopped. The logs are oddly quiet right now.

I'm also open to other ideas on how to monitor the node reboots, if the built-in one has some type of bug with it.

Also -- if anyone knows what log actually tracks the alerts executing out, that'd be interesting to look at. This would at least help me see if the alert is firing off, but I suspect it's just not picking up on that change in the database.

silverbacksays

Hey rboyer

Take a look at the dbo.AlertHistoryView under Views on your SQL server. That should show you if your email triggers have fired off successfully or not

NPM 12 upgrade reboot server alert is not working - SolarWinds Worldwide, LLC. Help and Support

It turns out, I just needed to reboot my server, but here's the steps I executed against my server.

1. Run a repair on Core Services in control panel.

2. After the Configuration Wizard runs, stop all services.

3. Open Database Manager.

Run queries against the database:

delete from dbo.PendingNotifications

delete from dbo.SubscriptionTags

delete from dbo.Subscriptions

4.Reboot the server.

The funny thing is, I repaired the core service due to an assembly missing in the windows directory, which lead me down this path to begin with.

You can also follow the below link, same info is given there.

-R

It is my understanding from support that those 3 queries are only for multi-polling engine deployments - is this true or false?

Hey orioncrack,

I'm not sure where that information came from in support, but I've been told at least twice through support cases to run it against my single polling environment. (SAM/NPM/WPM) I do not run additional pollers in my environment and this did fix my problem.

-R

I see those 3 queries everywhere, why is this code the swiss army knife of fixing SolarWinds?

Came here looking for this exact answer, thanks to everyone in here.

ArjenBrands

I use the very simple, but effective sql query as workaround trigger condition:

SELECT * FROM [dbo].[Nodes]

where (DateDiff(mi, LastBoot, getdate()) < (3))

The evaluation frequency of the alert is 1 minute and the reset trigger is reset after 3 minutes

Going to bump this back to the top... anyone else have this problem return? I just did now and had to do these exact same steps to resolve it. I really don't want to have to do this every two months, has anyone opened a support case or received any info from SolarWinds regarding this? If it's a "thing" in NPM 12 now that I need to remember to do or will be forced to do that won't be much fun.

One thing I did do recently was that within the past 48 hours I rebooted the database server because it was chewing through memory... and that's with the recommended 128 GB assigned.

bmoore1

Ding Ding Ding!

I am having this same issue. I have 3 separate tickets on this exact problem with support since we made the plunge to NPM12. The last moron SolarWinds Support Rep told me to install 12.1, it will fix it.

So, I reviewed the changelog for 12.1 and nowhere does it mention Reboot alerts. I followed up with the REP and he requests to close the case.

The quick "fix" is always to upgrade.

In the middle of such silliness now with 2 clients.

bmoore -- Yep, I think mine started when I went from 11.5 to 12, which started awhile back since i'm an early adopter of their releases. I guess I like to put my head in the fire for some reason.

I think their tables get mucked up when you bounce the SAM/NPM server and their queued alerts don't unclog from the tables in the database.

I've updated recently and guess what? It's still broken.

I think the gremlins may go deeper than we think on this issue.

Orion Platform 2017.1

WPM 2.2.1

VIM 7.1.0

NetPath 1.1.0

DPAIM 11.0.0

NPM 12.1

QoE 2.3, CloudMonitoring 1.0.0

SAM 6.4.0

I've not tried the SQL based alert, which may be a decent work-around, but I've not tried it in my environment. It sucks since I have over a 1000 nodes loaded in my datacenter, so it's a PITA to bring down SAM during the day to fix this and it certainly doesn't qualify to get my attention in a maintenance window. (I'd rather throw it out the window, lol)

The reboot alert is fairly critical piece of information to have during the day and when I do finally clear it out, I get a flood of false positives, so i'm not sure which is worse at this point.

-R

That sucks that it's still an issue after the upgrade. I was eyeing the new NPM and SAM but now I have the primary, and three APEs so it's not going to be as easy as things used to be.

Waiting for the next go around with the software when this pops back up again. So far it's still working for me but I know it's just a matter of time...

Here we go again, just had to do all this good stuff again today. Got this thread bookmarked at this point for the next time this happens. Shame to hear so many people having problems with the upgrade right now or I'd just do that even though it looks like there's no guarantee that'll fix this, either.

mbird I submitted possibly my third ticket (Case # 1150965) to get this post some community feedback. This issue originated in July of 2016 from their KB's last updated. I feel the issue began with the UI refresh, if my memory serves me right here. The problem does get quite annoying at least with my server, it stops only when I reboot my SAM/NPM server. If I do not touch my server, it's fine, so there must be something that becomes unglued when the server is rebooted.

aLTeReGo Your feedback is pretty sound -- if you have some insight on what we can do here to better the community, I'm all ears.

I'm a little sordid with the situation, since my division is in dissolution and will have an end term date by the end of this year. I do wish to see this issue addressed and not ignored though. It does not show well with Solarwinds' support on passing off the workaround with no follow-up to their customers on a true fix to the issue.

I think your timing is about right. I had the problem late last year for the first time based on my post in this thread. I do believe that was right around when I upgraded.

I however have not rebooted the primary poller since the last time this happened, so it wasn't a reboot that caused the problem (yet again) for us.