NetXMS Support Forum

В configuration -> Server configuration настроить параметры SMTP.*
Нужно учитывать, что сейчас на поддерживается шифрование/аутентификация, если SMTP сервер это требует то на машине где стоит netxms нужно поставить и настроить какой-нибудь SMTP relay.

В configuration -> Action Configuration настроить действие.

В правилах EPP для которых нужно отсылать сообщения добавить действие, настроенное выше.

А в конфиге агента параметр Action= как прописан?

You need either SNMP daemon or netxms agent installed on the machines that have several NICs - then netxms will understand that all NICs belong to one machine. This will also solve Unknown network interface issue.

Once you have a node that is accessible via SNMP or netxms agent you can right-click on interface and select "Create data collection items". There you will be able to create the data collection items (DCIs) that will collect numeric data on the traffic. Then you can right-click node and select Last values - the data will be there.

Replicated this on Ubuntu 18 so looks to be a bug. Created an issue, you can monitor resolution there: https://track.radensolutions.com/issue/NX-2170

In the properties of node, which you are trying to poll under Communications there is "Primary host name". This is the address of agent to which netxms server is connecting. Take this address and add it to MasterServer line of agent config after comma.

Your debug approach looks to be correct. There also could be also some information under "nc" tag.
Is the actual .py script executed?

I've tried like this, it's working:
Command=/bin/bash -c "/bin/echo ${recipient} ${text} >> /tmp/sss.txt"

What exact version of netxms are you using?

There was a bug recently that upgrade command was actually not executed.
The upgrade however is uploaded.
You can create agent action that executes CMD.EXE /C "C:\NetXMS\var\nxagent-3.9.334-x64.exe" /VERYSILENT /SUPPRESSMSGBOXES /LOG /FORCECLOSEAPPLICATIONS /NORESTART
and it should do the job.
You can create agent configuration policy in a template and apply this template to your nodes - this way you can distribute this action to your agents.

If you install netxms agent on these machines (and add the address of your netxms server to MasterServer or ControlServers or Servers in agent configuration) then netxms would be able to get information about network interfaces. The other option is if SNMP daemon is running on the machines.
Without that netxms can only see that there is a machine and can ping it to check if it's running. But it can not determine that two addresses belong to same machine.

Most probably this means, that the address of your netxms server is not in MasterServers line of agent configuration.

If the agent is on the same machine as the server, the default configuration for the agent includes MasterServers=127.0.0.1, but the node that is automatically created refers to external address of the machine.

Maintenance mode only suppresses event generation, but data collection and status propagation is still happening. If the nodes are temporarily decommissioned, you can put the to unmanaged state (node context menu - Object-Unmanage) - then they will stop affecting parent container status.

Changing of interface expected state is possible from NXSL script. We currently do not have option to keep NSXL scripts in templates, so you have to use some sort of global script for this.
Hook::ConfigurationPoll is a good place - it's a standard hook script that is executed on each configuration poll.

A simple example is this:
interfaces = GetNodeInterfaces($node);
foreach(i : interfaces)
{
SetInterfaceExpectedState(i, "IGNORE");
}

but you need a bit more advanced script. You can check, if node is assigned to particular template (in this case node becomes a child object of the template, so you can use e.g. $node->isDirectParent(template_object) method to check this).

You can use mapping tables to keep list of interfaces that should be in "UP" or "DOWN" state if you need to.

Please inform, if you need more information or examples.

I did not test this, but probably it does not support format with \ in the database address field. Try just IP address there.

Anyways, all that the installation wizard does is creating C:\NetXMS\etc\netxmsd.conf and launching
nxdbmgr init
(which is located in C:\NetXMS\bin).

So you can adjust the .conf file and run nxdbmgr init manually. When it succeeds, you can start Netxms Core service.

Do you actually have MS SQL database installed?

It should work the same as for regular Postgres and other databases. There were performance fixes recently, so upgrading to the most recent versions is recommended.

Nxdbmgr requires to stop netxmsd, so you may prefer backup by some postgres means instead.

Yes, sure, if NetXMS Core service is stopped, it's won't poll anything.