shuth · SolarWinds Consultant · ✭✭✭✭✭

If you go into the View Map mode, when you click on a link a window should appear on the right showing the connection details. The screenshot below is from the demo website.

Next time you see a red line, click it and look at the right side for the interface details. Looking at your screenshots I'd say it's either errors or discards on the interface breaching the threshold.

If you get stuck, you can also search in the SAM templates for other Linux component monitors and check those scripts for more examples.

Ahh I didn't see the scroll down. As I said, I'm not 100% across the scripting side of things but I threw the following together and it tests OK in SAM. #!/bin/bash # test if port is available if echo | nc -v x.x.x.x XXXX; then printf "Statistic: 0\n" printf "Message: it is up\n" exit 0 else printf "Statistic: 1\n" printf…

Found the article I was originally trying to add in: documentation.solarwinds.com/.../sam-script-text-output.htm

The Linux script component monitor is expecting a Statistic output with optional corresponding Message. So you may output Statistic and Message, or Statistic.1, Statistic.2, Message.1, Message.2, etc. if you need to output multiple fields/statistics.…

You could do it in a custom table with HTML, but I don't think it does pages so you'd have one big list of nodes... hence going back to Custom Query but you're limited to 1 icon per column AFAIK. 1. Add a Custom Table widget 2. Add a Datasource - selection Method = Advanced Query (SWQL) SELECT N.Caption AS [Node] ,…

I'm not sure that is possible. Even if you try to create a Custom Table report, if you choose to show the Status Icon it hides the option to show the Vendor Icon (and vice versa). I think you might need a column for either the status or vendor. You can have a blank field and assign the icon but it's still a bunch of wasted…

If you're not migrating servers or databases and just upgrading then it's fairly straightforward. Extra servers like additional polling engines, high availability, etc add complexity but without knowing what you've got or if you're doing an online or offline upgrade it's hard to say. You can find all the prerequisites and…

I have used this feature and I'm lost in this thread - it seems like arguing over semantics. The wording for the setting could probably change from "Repeat this action every X minutes until the alert is acknowledged" to "Repeat this action every X minutes". It boils down to: - Click the repeat action option if you want the…

I think you may need to use the Custom Table Poller Current Status condition instead of Custom Node Poller. Then you will be able to filter by the RowID and other values.

Best guess is polling the VM directly vs polling the host - Peak CPU Load is from polling the node with WMI/Agent/SNMP - Maximum CPU Load is from polling the VM host and is what the host reports the VM is using

I have a system running 2023.4.2 and seeing the same thing. My lab is running 2024.1 and is not. The 2024.1 release notes contains a reference to fixing this. documentation.solarwinds.com/.../solarwinds_platform_2024-1_release_notes.htm 01334092, 01467443The OrionWeb.log file is no longer filled with warnings about a…

Not exactly what you're after but a workaround would be to run a Network Discovery against the nodes you want to enable Asset Inventory. Untick all of the interfaces, volumes, AppInsight apps that are detected and "re-import" the nodes. It shouldn't re-add the node as it already exists but will enable the CPU/memory…

I would enable debugging on that application monitor and review the debug log.

I can't recall the exact table but one of the Orion.VIM.?? tables in SWQL contains the link between the NodeID and HostID. You could probably put together a query that gives you a list of Nodes where the NodeID is NOT in that table.

I've done something similar in the past with by adding in the cluster IP as a monitored node. Server 1 - Monitor the services but mark them as Not Running instead of Down. Server 2 - Monitor the services but mark them as Not Running instead of Down. Cluster IP - Monitor the services, mark as Down You can use the monitors…

You can also change the custom poller in the UnDP tool to enumerate the values under the Advanced Options. Edit: Looks like that OID already has the values listed but you need to change the format from None to Enumeration.

If you're just after the links, I'd add the "User Links" widget. Otherwise, rough idea but potentially something like. SELECT DISTINCT Studio , 'Dashboard Link' AS [Dashboard] , CASE WHEN Studio = 'ABC' THEN '/apps/platform/dashboard/1' WHEN Studio = 'DEF' THEN '/apps/platform/dashboard/2' WHEN Studio = 'GHI' THEN…

Can confirm most of the support articles I've tried to access today have been broken as well.

You need to tell it to group by the other fields, otherwise it is counting the total across all your groups. This will show you all the combinations of ID/GroupCity/GroupState but should give you all the groups because the ContainerID will be unique. SELECT TOP 1000 c.ContainerID, c.CustomProperties.GroupCity,…

If you want to filter out certain databases/servers, you edit the data source. If the fields you want to filter by aren't in the basic dropdown, you can use the Advanced Selector option to find the metric/field. When it comes to adding the extra fields to the report, when adding the columns you can change the object. E.g.…

I should have grabbed a couple of the Astro Geek and IT Kong shirts before they went out of stock... mine are several years old now.

If using the Web Report GUI, if you create a Custom Table you can use the data source "AppInsight for SQL: Database" and select the Recovery Model from there. Filter by server/database/whatever in the source, and you can also add the database name, server name, and any other values from there.

From what I have seen, I believe this is the behaviour when a node gets added to the system. * If using SNMP/WMI, it will use the detected System Name * If no system name, it will do a reverse lookup of the IP address against DNS * If no DNS entry, it will use the IP address You could probably do a bulk update against the…

I had a client with a similar issue when they were upgraded to 2023.3.1. They use the All Nodes widget with queries pulling in NodeIDs from groups/containers but with SQL. Post-upgrade all of the widgets were broken with errors about the filter. Turns out that on top of converting the query from SQL to SWQL, I found that…

Due to having multiple IP addresses on the same NIC, it comes down to how Windows chooses which IP to use and that depends on the IP address of your default gateway, the VIP, and the active server IP address. Basically, the IP addresses are converted to binary and the IP with highest matching bits to the gateway will be…

If you have more than 100 Error messages, you probably need to put an "ORDER BY MessageDateTime DESC" on the end of the query. Otherwise you are grabbing the first 100 messages in the table. With the ORDER BY on the end, you will grab the latest 100 messages. Potentially as well, you may need a "TOLOCAL(MessageDateTime) AS…

I'm using SWQL Studio and can see them in there. I use an account with Admin role in SW which has access to everything so don't know off the top of my head what permissions are required - maybe Alert management, maybe full admin. Also, ProgramPath isn't the name of the column. Per my earlier screenshot, the…

The data should be in the Orion.ActionsProperties SWIS table, and can filter by the PropertyName = 'ProgramPath'. * Orion.AlertConfigurations - the alerts and corresponding AlertID * Orion.ActionsAssignment - find the actionID #s by matching AlertID = ParentID * Orion.Actions - can see the basic actions information (title,…