NetXMS Forum

Ill add some more info to this. If someone wants a more readable status of the service poll you should under General change Data type to "String" Then you go to Transformation and paste this code: switch ( $1 ) { case "0": return "Running"; case "1": return "Paused"; case "2": return "Start pending"; case "3": return "Pause pending"; case "4": return "Continue pending"; case "5": return "Stop pending"; case "6": return "Stopped"; case "255": return "Unable to get service state"; default: return "INVALID VALUE (".$1.")"; } If you want to get notified via lets say e-mail when a service stops you will need to look at Thresholds, Event processing policy and also Actions configuration. [image: Skärmavbild 2021-10-27 kl. 21.47.23.png] [image: Skärmavbild 2021-10-27 kl. 21.47.43.png] [image: Skärmavbild 2021-10-27 kl. 21.48.00.png]

Thank you for the detailed response. This is exactly what I needed.

When looking at the Alarm Browser, Summary Tables, and probably others, you can right click and choose Object Details. It would be helpful if the Object was selected in the Objects tree to easily identify which customer Container the Object is located.

The only option I see is to set custom attributes with maintenance entrance time on EnterMaintenance($node) function execution. Create a script that will go through all nodes every minute and will check the custom attribute with maintenance start and will initiate maintenance leave in case if the note is in maintenance more than an hour.

Exactly what I needed. Thank you!

Hi, you can use function AgentExecuteAction to execute action on remote agent. So script which will cause remote node to restart will looks like this: AgentExecuteAction($node, "System.Restart"); you can create scheduled task which will call this script. Best regards, Victor

You can use /CENTRALCONFIG or /LOCALCONFIG to set this (defaults to /LOCALCONFIG). We use "/VERYSILENT /DIR="C:\Program Files\NetXMS" /CENTRALCONFIG /SERVER=hostname_of_netxms_server /NOSUBAGENT=winperf" to do a completely blank install, and then serve the config from the server. Peter

Hi, you can also add any missing configuration parameter directly from UI (there is option "add parameter" in server configuration view) or using nxdbmgr: nxdbmgr set AgentTunnelListenPort 443 Best regards, Victor

That was the issue. Also, if Agent Cache Mode is set, Force DCI poll does not work. PS Thank you for the excellent audit logs so I could find out who and when changed the default setting for the cache mode.

Install the RabbitMQ Management Plugin https://www.rabbitmq.com/management.html In the nxagentd.conf of the node I want to monitor RabbitMQ I put: ExternalParametersProvider = "C:\Windows\system32\WindowsPowerShell\v1.0\powershell.exe" "C:\NetXMS\var\rabbit.ps1":60 Restart the agent This gives you a Rabbit parameter. You need to poll the node for the config first to see the parameter. Right click the node and see attached screen shots on how to poll and verify parameter is available I used the attached powershell script. Make changes to $username, $password, $rabbithost, $vhost (if not using default of / (%2F)), $queue $workingfile gives back a table of all the statistics for the queue The issue was that the default vhost in rabbit is a '/' which needs coded as %2F. It runs fine in the powershell ISE and was even writing out a debug text file with the correct value. However, when run from the command line, it fails with a "not found" error because the '/' is not interpreted correctly. The function at the top (which is not my own) cleans up the URI correctly and now I get the value back in NetXMS. rabbit.ps1 [image: agent_poll_configuration.PNG] [image: agent_supported_parameters.PNG]

Hi, it's a bug, already fixed in development branch. Best regards, Victor

Victor replied on a different thread: Hi, default templates re-imported from file system on server startup. You can disable this behavior by setting server configuration parameter ImportConfigurationOnStartup to 0. Best regards, Victor

Got it thank you. I'll put this over on my other thread that I started.

I was up against my upcoming maintenance window and decided to put in something quick and dirty based on your suggestion. I separated the email and agent action Event Processing Policies I added an unmanaged node, "Patch". Set a custom attribute on the "Patch" node, "PatchTime". If I set "PatchTime" on node "Patch" to something other than 0, that means I am in maintenance mode and do not want to see email alerts In the email event processing policies I added this to the Filtering Script: $patch = FindNodeObject($node, "Patch"); $patchtime = GetCustomAttribute($patch, "PatchTime"); if ($patchtime == 0) return true; else return false; I can manually change the value easy enough and this meets my needs on a basic level. However, I am planning on taking this further and walking from the topmost parent down to the node with a similar (or same) custom attribute setup. That way I can be granular with my "maintenance light" mode. Open to any further suggestions for improvements/functionality.

Registered as feature request: https://dev.raden.solutions/issues/1354

I got it working. One thing that kept tripping me up is that when I would make a change to the discovery script and then Force DCI poll, it would say that it was unsupported. So I would make more changes and that would be unsupported. I probably had it working earlier and just needed to be patient for NetXMS to do it's thing. I had to still see if it was Windows 2003 because newer versions use both services but only the MpsSvc is the firewall. [image: Capture.JPG]