designerfx · observability architect · ✭✭✭✭✭

it's not illogical to run the config wizard after you finish vs after each upgrade - because the hardware and config should theoretically be the same. Maybe the product upgrade advisor could one day acknowledge or say something like "upgrading 3 modules? run CONFIG WIZARD at the end!" I do say this having seen this just…

I do exactly that as well. Save your visio as a jpg and done. I do find I prefer large status icons over small ones, depending on what visio stencils people were using. 1400x800 sounds about right, though.

This was the first idea that came in my mind also, vendor for category under managed nodes and then sort by yes/no for NCM licensed.

Sorry, I left my previous job using solarwinds to find a new job using solarwinds - in the middle I didn't end up logging into thwack! Glad this was resolved.

Just for my own understanding of EOC which I've never worked with or had deployed yet, do people normally do that? The putting EOC on the SQL server part?

not only that leratob​ but if you scroll to the right you'll get details on what needs to be done to fix those.

Yeah, this is the purpose as explicitly laid out by jangliss . They cache info for example, in case you need to change something in what you *want* to poll but there isn't a change from the specific machine. As in, if you weren't polling memory and now you want to. If it's "a new module was added", then you don't want to…

I have been extremely unwilling to test out auto dependency. I thought once you turn it on it immediately makes the new dependencies. Does it work without doing that now or do you have to sort all the auto dependencies first?

It's a bit too locked down right now. Top indexes across our 100+ SQL servers all tend to go into alert at the same time and are completely suppressing our event log, and there's no way to even change the criteria.

Wait, people use the actual name of the object as a link? I tend to find that I can link to anything via the map and have no such issues, but I don't rely on the name. re: atlas on the wayside, more and more things are being linked into it. WPM's data for an example of something done in the last few months. Atlas isn't…

I never knew about this until I read this today, and I was looking for something exactly like this. Thanks!

yes, but that requires having the OID so that solarwinds can set it up and/or let cobrien​ have someone integrate it.

jiri.tomek how do you enable verbose and/or what to check for in the logs to really sort out performance issues/slowdowns within them?

I was thinking along the lines of rschroeder​ as well, even though there's no exact answer on this one.

surlyitjesus​ I talked to I think either aLTeReGo​ or rob.hock​ about this when I originally posted it, but I don't recall the response as it's been almost a year. Maybe they can get an answer from the Web/UI/UX folks.

Account limitations are designed to enable people to look at X or Y thing, but they're super inflexible so I wouldn't suggest relying on those.

Can the app pool handle a web garden approach aLTeReGo‌ ? Aka multiple worker processes instead of 1?

No sir. I never got any responses. Kinda dropped it, but unfortunately the risk of excess log storage remains. Maybe stevenwhunt​ would know who can answer this.

dodo123​ did you just indicate use port 80 (common for HTTP) and checkbox use HTTPS, which usually uses port 443? Was that default from adding the node? You probably need to fix that unless you're using HTTPS over port 80, which sounds unlikely especially if you haven't changed defaults on the blade.

This actually makes sense to me. I was under the impression this didn't work somehow from what jest4kicks​ and dmeiser​ were saying?

It's 2017, but I wanted to document this for lukepev​ : - InInterfaceDetailsLink should be - InInterfaceDetailsLinkAbsolute .

I also caught that this may be coming back going forward. There was a mention of an integrated search coming.

aLTeReGo so, what can be done to actually perform a SQL query that returns a column of values? Should I not even be using the Oracle Experience Monitor components for this and should I just go back to using powershell?

I realized this later on as I figured out the issue. You are correct. Thank you madhavan‌.

This is how I did things at my previous job. Put the SAM agent on each server in the cloud, use a proxy to send everything back through a single IP, monitor as if onsite. I think you covered everything, but SAM 6.4 now additionally polls cloud infra in the same way SCOM traditionally did - through the subscription. Orion…

I know, right? When are they going to ask experienced folks for UX feedback on NPM itself? I've got lots of areas that I think need some improvement.

Also as a matter of best practice; you are not going to realistically be able to respond to things sooner than that. One other thing to watch for is a solarwinds alert for "high packetoss", as that will pick up on things during fast polling if you wish to leverage it. Warning: it can be noisy.

Your issue sounds like possible VM overprovisioning. How many cores do the actual processors hosting your VM have? If you have either the same number or more vcpu's provisioned than the physical host you're going to have IO issues that can end up spiking CPU use substantially.

bluefunelemental‌‌ as I'm our SWQL guy and work with OP (self titled, haha), I ran this against another "down" volume in our inventory and most of them show Last Sync as today ~ 6 hours from now (future polling!) I'm assuming that's just a time conversion and more or less means it's been polled, but nothing showed a date…

My first thought was EOC just from the subject line, but that really only makes sense if a: you have enough things that more AP's isn't going to help or b: it's separate companies. If it's one company you really don't have a reason to split it as two installs (as opposed to having more APs aka additional pollers).