jest4kicks

This, this, oh god so much this. We're actually having some issues with the amount of polling we're doing, and our DBA team has indicated they really don't need 80% of the stats that AppInsight for SQL polls. I really don't like wasting polling capacity, and now we're a little stuck doing it.

Hey Shack, Just reading your comments and train of thought, I can tell you're running through the same sequence of attempts that I did. Tragedy + Time = Comedy? I wrote a thread about how we ultimately solved this. Cisco UCS trap-based monitoring Cliff notes: you need to configure your CIMCs to generate traps to…

Actually I can confirm that Support said this method of disabling AppInsight components was supported. (Ref: Case # 1011232) That said, we disabled several component monitors, and it completely broke the overall monitor. (Ref: Case # 1011378) So while you might be able to disable some components this way, there are…

Is this a Solarwinds-approved method for disabling the components? Our DBAs refuse to make any manual edits to a third-party database without specific approval from the vendor. Also, just pointing out, we'd want to disable 80 components of the monitor, and we have 35 instances of that monitor. That's 2,800 row edits. Even…

Answering my own question. Just talked to Support. Disabling components using this method is approved and they're planning on putting out a KB to reinforce that.

These OIDs actually have to be polled through the CIMC (Cisco's remote mgmt card, similar to a DRAC or iLO). The idea is that you add these Universal Device Pollers to them. While this will give you some information, I've found it incredibly tedious to try and build custom alerts around this MIB. Instead, I'm looking at…

We do it by Solarwinds grouping. I don't think I see anything in the Nodes table that corresponds to that. Do you know where it's stored and how to link-in that data?

My thoughts exactly. Not thrilled with Solarwinds' support line of troubleshooting right now.

I might be using the wrong term, so I'll just provide an example. We have multiple teams that we want to be able to provide Solarwinds access to. We don't want them to have to wade through all the other systems, so we've set view limitations on their access groups. Those limitations mean that, when they access Solarwinds,…

We haven't started using agents yet, so I'll have to look at that config and see if we can give it a shot. The case I opened with Solarwinds hasn't gotten very far. They're saying it could be a TLS/SSL protocol compatibility issue (not sure how that would be). According to them, SAM 6.2.x only supports TLS 1.0 or SSL (no…

That's another good thing to check, but unfortunately not applicable here. We're set for unlimited licenses and are currently using 1,185.

I did. Problem - AppInsight for SQL and passive DB copies Thanks for the help!

Hmm, that doesn't seem to work. I'm getting the following error when I did deeper. Msg 207, Level 16, State 1, Line 6 Invalid column name 'DeviceRole'. Msg 207, Level 16, State 1, Line 6 Invalid column name 'DeviceRole'. Is DeviceRole a custom property in your instance? I replaced it with a similar property that we use and…

No luck. Those values are all present as expected (IIS 8.5). The server is not part of a cluster.

Hmm, I was hoping to make it more presentable in the web view. Sounds like it'll have to be exported to excel. Ok, I'll keep working with it. What about the view limitation? I searched around and saw that view limitations seem to be a problem with custom queries. Is there any way around that?

*cough cough* Cisco UCS trap-based monitoring

Thanks for that info! Is there any method planned for bulk-unmanaging databases? Or perhaps improving AppInight's cluster-awareness?

Well that feature wasn't installed (though it begs the question of how the monitor initially worked without it)... however installing it doesn't seem to have addressed this issue. Is there any way to know what Solarwinds is looking for when it determines if IIS is available for monitoring on the server? I assume it has to…

What do you mean by "add them from the UCS manager's details screen"? I'm just looking at the UCS FI that we added as a node in solarwinds (technically the management name/IP, not a specific FI). I don't see anywhere to add nodes here (as opposed to a vmware host, for example, which lists additional nodes and allows you to…

Thanks, Dave. Barring confirmation that this type of delete won't create link or orphaned object issues, it's probably the safest path.

I'm with you. I definitely didn't score any points with my DBA when I told him it HAD to be SQL auth. 

Oh ok! Well there's a couple things I'd like to learn how to do differently. Any light you can shed is greatly appreciated! First, the report is producing one line per physical volume, with all the other details of the server duplicated between those lines. Is there a way to group the volumes together so that it just…

You can, but here's the solution we ended up using. (Your post on UCS UnDP was probably one of my previous attempts at a solution for this. Thanks for that!) Cisco UCS trap-based monitoring

This summarizes the issue perfectly. It makes me wonder if there is a date recorded when the volume is added, that could be referenced in the custom volume forecast query. (Smarter Volume Capacity Forecast SWQL alert ) Gonna look into that now, but anyone feel free to chime in if you know.

Yeah that would be awesome. In the meantime, what about using Set-swisobject? Am I using the command incorrectly? I don't understand why it doesn't like the URI.

Great answer. (And in line with what Alterego said, too) Thanks for the input!

Yup, that's what I tried. No luck. Thanks for clarifying, though!

There are some other OIDs we are currently polling for disk status, but I was hoping to avoid creating individual alerts for every possible hardware health OID. Monitoring faults should cover all hardware issues, but it doesn't look like there's a good way to do it without relying on traps.

This is in the Orion web console, alert management. Alerts > Manage Alerts > Add New Alert > Trigger Condition > I want to alert on: (scroll down to Custom SQL Alert) Also btw, did you know your username for the forums is your email address? Not sure if that was intentional, but you could end up discussing some…

That's a great write-up for basic SNMP trap-based alerting! That said, I'm not sure how well that simplistic approach would work for Cisco fault traps. The extra SQL language in our alert is designed to deal with how the CIMC sends multiple traps for a single fault. It filters out much of that noise while still reporting…