NetXMS Support Forum

netsvc should not be required for web services. It should be present in netxms-agent package, please check output of
dpkg-query -L netxms-agent
If its not there, what is the version of OS?

No, nxget, nxaction, nxwsget do not support tunnel connection - they attempt to connect to agent directly.

А какая точно версия NetXMS?

Можно на лету поднять ненадолго уровень дебага до 7 уровня:

nxadm -c "debug 7"

и попробовать отправить письмо. Потом опустить дебаг обратно

nxadm -c "debug 0"

Возможно, в логе будет что-то интересное.

Еще, как вариант, поснифить обмен с 25 портом на localhost с помощью Wireshark.

У меня не получилось повторить. Какая точно версия? Можете проверить на самой свежей 3.7?
Проявляется только в трансформационном скрипте, или если на ноде сделать Execute Server Script, то там тоже? Что именно в $1?

I've created an issue for this feature https://track.radensolutions.com/issue/NX-2003, but it might take quite a while until developers get to this.

Видимо что-то поменялось в языке. По документации SecondsToUptime принимает целый агрумент, так что можно округлить:

SecondsToUptime(round($1/100))

It's recommended to connect using mysql client, not 3rd party software, as heidisql is crashing here.
https://dev.mysql.com/doc/refman/5.7/en/mysql.html

It all looks like some issue with MySQL, possibly the database is corrupted. May be you can find some useful information here: https://stackoverflow.com/questions/25039927/database-corruption-with-mariadb-table-doesnt-exist-in-engine

If you have backup, I would suggest to restore from it.
Also, we generally find Postgres more reliable (if you decide to go this way, nxdbmgr allows to migrate between database engines).

A possible way to debug here is nxwsget command line app that is included with the server. This way you can isolate, whether the problem is on agent or on the server side. E.g.:
nxwsget 127.0.0.1 http://api.open-notify.org/astros.json number

Can you check in MySQL if "nodes" table exists?

Did you also upgrade MySQL itself? Can you check, if there are any errors in MySQL log file?

One possible option is that there is an issue with access right to these nodes. Try enabling "system" user
nxdbmgr reset-system-account
(it's password will be "netxms"). This account has access rights to all objects in the system. Could be that the nodes will be visible (may be in the root of object tree).

Для последней версии nxmc требуется 11 java (это требование фреймворка, на котором собирается nxmc)

You can modify instance name and display name in instance discovery filter script by supplying them in array that script returns. Just open "Hint" in filter script editor, it has the necessary information. Here's an example script:

m = $1 match "(.+)\[.+\]";
if (m) {
  return( %(true, m[1]) );
}
else
{
  return false;
}

First line matches instance against a regex. The regex has one capture group which returns anything that is prior to [.
If regex got matched we can get this capture group as first element of array m.

Just in case, it's possible to read port forwarding database from NXSL - it's available in Topology.SwitchForwardingDatabase internal table. There's no access to L3 routing information yet, but it's planned to add https://track.radensolutions.com/issue/NX-1999

It's not fully implemented. It adds System.Execute action (and also System.ExecuteInAllSessions on Windows), but System.Execute does nothing.
Updated docs.

What device is that? Is it polled by SNMP or agent?

Выглядит все вроде бы нормально.
В Monitor->Events или Event Log стоит проверить, что событие CPU_LOAD_NORMAL действительно генерится.
В EPP нет нигде выше правила с галочкой Stop processing, которое срабатывает на CPU_LOAD_NORMAL?