brscott · Custom Property Aficionado · ✭✭✭✭✭

A couple of things here. a) You can alert on the group, such that if anything in the group goes "red" then it generates an alert, but on the group, not the individual node. Generally this is not what people want. b) There is an alert category called "group member" that you can alert on, but when you try this you'll learn…

Once you've created the true feature request. This discussions is still useful to you, assuming that support didn't have a fix. So, step 2 is developing a work around yourself. What I read from your description is that you have a script that outputs a large number of data points. Lets make up a number, say 200. This is…

There are probably assumptions all over in the code about the max 10 statistics, so support is probably going to tell you there is not a "fix" for it. If that is what their response is, then step 1 is to create a feature request. This looks like a discussion. From the Top Menu -> Product Forums -> Server and Application…

Also to answer the question. Yes, multiple different alerts will fire for the same node or sam monitor.

When you copy an alert, it's disabled by default. Check to make sure it's enabled.

The base answer is No. Not the way you are thinking. Most handle this inside the ticketing system, yes it creates two tickets, but this is generally considered good. Either way, the tech working on it needs to get notified that the system went from warning to critical levels. I'm guessing that you are using simple "Volume"…

Just to be clear. You only need the SDK to create the groups for NEW cities. You can already populate the groups dynamically without the API. If you are not adding and deleting cities often, then maybe look at “Add Dynamic Query” on the same screen as you would manually add each node to the group. I just want to make this…

a single application listening on 200 ports is not really that common. I mean unless you are running 200 full xwindows desktops over VNC... then yeah. Seriously though, it might be a lot easier to run a single script monitor that outputs something like: statistic: 150 message: 150 ports listening, 50 ports down

Ok, next step. Assign the template to the node, even though it's not working quite right. Edit the monitor (or I guess you could do this in the template), and turn Debug: On. And do a "poll now". Then on the node, go to /opt/SolarWinds/Agent/bin/appdata/Logs/ApplicationLogs/AppId<blah> See what the log says.

There are a few ways to do this. 1) Create a node for each interface, see Craig's response above. 2) Add a widget called "Down Interfaces", then edit the widget to filter down to what you want. 3) Add a "Custom Query" widget, and form the SQL/SWQL yourself. This is the most difficult option, but also the most flexible. 4)…

In the email trigger settings, check "Time of Day" and "Execution Settings". Or you added an escalation level and the email alert is a level deep with a delay, but you would see that right on the "trigger actions" tab of editing the alert. It's not somewhat hidden like the other two I mentioned.

Nevermind..... I am looking right at the option on my edit template screen "Count statistic as difference" = false. Now, why is it not working.

Please let us know if they have a fix for you. I have the same issue here, I currently only have 4 networks scanning across 3 polling engines, because turning on several hundred would kill it. I'm pretty sure that it's something in windows because as you mentioned, pings and tracert from the command line no longer work. It…

Ultimate answer: It depends, but probably doesn't matter much either way. Personal opinion: #1 for the main poller. Reason: Solarwinds can really use all of the cores and in an ideal environment to estimate I would do the math as (10x2.6 = 26Ghz) vs (4x3.3 = 13.2Ghz) #2 for the SQL server. Reason: I "think" (I'm not the…

There are a few of ways. 1) The easiest way, is to set the warning a critical levels on the volumes. The problem here is that it must be expressed in percent of the volume, and that percent number must be a whole number. So if your tolerances as less than 1% then this will not work. For example if you have a 2GB volume and…

No there is nothing to send traps, but that would be AMAZING! Now there are plenty of things that can send a trap with whatever information you want... but if you knew everything you wanted to put in the trap, then you would not have misconfigured the alert in the first place... Someone needs to write a utility: 1) Select…

I noticed this on some machines monitored by the Orion agent. It started during the time change. I unmanaged the node, waited 30 seconds, then re-managed the node.

Simple answer: Write-output "Statistic: $Days" Full Answer: SAM Script Component Monitors - Everything you need to know

The Solarwinds Orion mobile web interface is really good. I would not allow it on the Internet, but we we have a team that works on web apps and a team that worked on Mobil device management, and they got together and setup Orion on a per-app-vpn and some other magic. Sorry I don't have the details on that, I just wanted…

1) You have to specify each custom property you want to display. I mean there is almost always a way... but no, that would be a horrific swql. 2) the search swql is a complete and independent query. So the custom SWQL would look like: select n.NodeID, n.NodeName, n.CustomProperties.something,…

No, I had the same issue. Everything is working as expected. Your alert is perfect, the condition is met, then the next step is to see if the node is muted, which it is, so the alert doesn't fire. I had to end up writing a script using the API to check for this and create tickets in our helpdesk system. It would be nice to…

I regularly have between 20 and 35 people connecting to a single additional web server, and I do sometimes notice things getting slow. I really don't want to put any of that load back on the primary poller, so I will probably buy another APE next year.

We are going to need more information. There are a lot of different setups for multiple network connections. The generic answer is: you have a way to know which internet connection is having problems.... monitor it. Router interface status, interface bandwidth drops to 0, ping ip address of the providers side of the…

There is a setting in the application monitor, expand Advanced section (just under polling timeout). In there can you can change the "Preferred Polling Method" to Agent or Agentless. Though on this specific monitor, I have never tried agentless, it should work.

I'm not entirely sure how you got "Server" to print out. Some of that would depend on what you are using as a shell. Anyway. Lets go through some suggestions. 1) Specify what executable you want to run the script with on the "Command Line:" field. Instead of "${SCRIPT}" use "/usr/bin/perl ${SCRIPT} or "/bin/bash ${SCRIPT}"…

Optimally, do as Serena suggested and adjust the thresholds so it only triggers when needed. BUT, I feel the need to answer your question even though I don't think it's the best solution. You can change the "Reset Condition" to "Reset this alert automatically after 24 Hours". That will leave the alert out there for 24…

Try changing to "I want to alert on" to Component. From the component you can access the application that it's a part of. It's easier to go that way.

I would put all of the nodes for a jurisdiction in a group. Then you can filter lots of things on that group. Alerts, world map, etc. Granted, it's usually not easy to filter things on group membership, but those can be more specific questions in another thread. if I understand your boundaries correctly with the…

We talked about this in another thread: What nodes don't have their disk volumes monitored? That thread only had a limited solution that works on virtual machines.

I am also on 2019.2.HF1, and that seemed to be the most stable agents so far. Check to make sure the agent version is 2.2.0.860. It should be if you have automatic update enabled. Spot checking a few doesn't show anything near 2.6G, mine are around 73M to 88M. I know some of this is the plugins it has installed, but I've…