NetXMS Support Forum

So far NetXMS manages to have backward compatibility, so server can work with previous version agents.

Can you show the full script. Original error message says the error is in line 8, but it's not clear what line is it.

Is EnableSNMPProxy=yes set in agent's configuration file?

If this is set, try collecting agent log on debug level 6 at the moment when you perform configuration poll from the server.

This is now fixed in 3.8.382

This was addressed in Telegram, let's copy this info here:

execute manually the following statements:
ALTER TABLE alarms ADD rule_description varchar(255);
(may file if alter was already done by nxdbmgr and didn't rolled back - that's ok then

UPDATE metadata SET var_value='17' WHERE var_name='SchemaVersionMinor';

server should start after that

Из "готовых" девайсов поддерживается протокол NTCB (https://navtelecom.ru/ru/).
Также есть поддержка USB/Serial GPS приемников агентом NetXMS, т.е. оно будет работать на каком-нибудь Raspberry Pi.
И есть Android агент, он также поддерживает GPS.

The reason could be that caching_sha2_password, let me cite a paragraph from the docs:

Bundled MySQL database drive does not support caching_sha2_password authentication which is default for MySQL starting from version 8.  Either select Legacy Authentication Method  when  installing  MySQL,  or  use  database  driver  installed  along  with  MySQL.  Databasedriver gets installed when installing MySQL with Server-only option,  however these two folders should be included into system path:C:\Program Files\MySQL\MySQL Server 8.0\lib C:\Program Files\MySQL\MySQL Server 8.0\bin.

Currently CreateSNMP transport does not support this. I've created an issue, also there's no particular plan when this could be implemented https://track.radensolutions.com/issue/NX-2062

Meanwhile you can try creating multiple nodes with the same IP address, setting checkbox "Communication through external gateway" (this is needed so that system ignores same IP address for these nodes). Each node would have specific community string set. To consolidate all monitored values onto one node you can configure "push" DCIs on that one node and use PushDCIData() from transformation script in DCIs of the additional nodes.

If your device returns list of interfaces for all of these community strings, then interfaces with same IP address would be created. This should not cause much troubles, but you can handle this in Hook::CreateInterface to avoid interface creation for the additional nodes.

The code is correct for nodes. But configuration poll could happen for other objects, e.g. clusters. In that case there is $object variable, but $node is not available.
You can change the 2-nd line to:
if (classof($object) == "Node" && $node->driver  == "MIKROTIK")

If osquery supports ODBC, then it could be accessed that was as dbquery sybagent supports ODBC.

Otherwise it could be technically possible to develop netxms database driver that would support osquery.

What is the value of AllowedCiphers in Configuration -> Server configuration?

Did you install from packager or compiled? Can you set DebugLevel=6 in server configuration file, start the server, wait for it to crash and share last few lines of the log?

It's saying "Refer to the logs for more information". Can you check if jira is producing any logs?

If DCI has node name in "Template" column this means that this DCI was created by instance discovery. Somewhere on the node there should be prototype DCI, which defines how instance discovery is happening.

e.g. in the attache image DCI with ID 4957 is instance prototype. It's coming from Generic UNIX template. Other DCI are created with instance discovery based on this DCI.

The script in your example returns only one alarm for each key. E.g. if you have several node with NODE_DOWN alarm, only one will be returned.

Can you give some information on what you are trying to achieve as the final result?