NetXMS Support Forum

Hi,

this is not an error, just message dump with system event configuration being transmitted to client. If you have SMTP issues, you should look for lines with word SMTP.

Best regards,
Victor

Hi,

your initial configuration is correct, seems to be a bug in a server. What version you are using?

As for alarms auto-resolve, you should set alarm key when create alarm, and use "resolve alarm" or "terminate alarm" action when processing recovery event with matching key. You can use default rules for node down/node up as an example.

Best regards,
Victor

It will need changes in server code - currently template objects are not exposed to NXSL.

Best regards,
Victor

Вроде как все проблемы решены. Здесь https://cloud.netxms.org/index.php/s/tm76wuhivyp0WNY можно скачать пакет с патчами (ну и на тестовой системе 2.0.5 сейчас собран).

Скорее всего поменялись какие-то поля в протоколе и старая консоль не принимает список параметров как надо. Надо разбираться с кодом старой клиентской библиотеки.

С nxalarm - также скорее всего баг в коде. Если теряются все обновления, то проблема скорее всего связана с изменениями в протоколе, если только некоторые - баг в обработке серверных уведомлений. Если нужен именно speech API, то мы можем попробовать добавить такой функционал в Java консоль.

Hi,

actually, SSH data collection will be introduced in 2.1-M1 (it is already implemented in development branch). Right now you can try to build agent with SSH support (you'll need libssh development package installed) and use SSH subagent. It provides single parameter called SSH.Command with the following arguments:

SSH.Command(host,login,password,command).

For example, to get load average from AIX machine:

SSH.Command(10.10.1.1,root,password,"uptime | cut -d : -f 4 | cut -d , -f 1")

And server version 2.1 will hide these details and allow you to create DCIs with SSH source and convert them into requests to appropriate agents.

Best regards,
Victor

Да, пожалуй доступ к системе единственный действенный способ.

Hi,

so there is another bug - node is not marked as changed when instance DCI is deleted and so changes not saved in database unless other changes were made. I'll fix that in next release. And workaround is to change something on the node to trigger save.

Best regards,
Victor

Repeat events for table thresholds not implemented yet.

Best regards,
Victor

Hi,

I'll try to test it in a next few days. Can you please try to delete base DCI, check that instance DCIs disappear in GUI, and check if they are in database (table "items"). You can select all DCIs for node with query


SELECT * FROM items WHERE node_id=?;


(replace ? with actual node ID). If they will be there, try to change something in node properties (rename it for example), and after some time check DB again.

Best regards,
Victor

Hi,

error 15027 means "the message resource is present but the message is not found in the string/message table". So application logs message with code that is not present in message table. It looks like a bug in some application.

Best regards,
Victor

Hi,

currently not. You can add feature request for that.

Best regards,
Victor

Hi,

server connects to agents (on port 4700 by default, but it can be changed). You can also have one agent on remote site working as a proxy for other agents - in that case you only need to open port 4700 to one machine on remote site from management server. Agent can also be used as proxy for SNMP, so you can monitor network devices via agent without direct SNMP requests from management server.

Best regards,
Victor

DSN должен быть системным. При проверке DSN, используете пользователя sa и пароль?

Попробуйте приложеную версию драйвера.