NetXMS Support Forum

Hi,

those errors should not cause service to not start. Please try to run netxmsd from shell with -D9 option and post last lines of output.

Best regards,
Victor

Hi,

yes, if you want average per minute then you put 60 as sample count. You do not need to collect it from agent every second - agent will return average for last 60 seconds whenever value is requested.

Best regards,
Victor

Try this filter script:


snmp = CreateSNMPTransport($node);
return SNMPGet(snmp, ".1.3.6.1.2.1.31.1.1.1.6." . $1) != null;


It will filter out all instances where SNMP-GET on .1.3.6.1.2.1.31.1.1.1.6.{instance} fails.

Best regards,
Victor

Did you add

EnableTCPProxy = yes

to config of proxy agent? Also, server has to be listed as master server on a proxy.

Best regards,
Victor

Hi,

one possible option is to create fake nodes without IP address to represent different units and use status DCIs to set unit status. I will also think about implementing some sort of built-in mechanism for representing units of distributed stack as that seems to be quite common task. If you wish you can follow on this ticket: https://track.radensolutions.com/issue/NX-1657

Best regards,
Victor

Actually, nxpush/nxapush was not designed with "update" option in mind. Server expects that each new push will contain new data. If you don't mind please create feature request in our tracker for "update" functionality for push DCIs.

Best regards,
Victor

Hi,

there is an option in node communication settings called "Prevent automatic SNMP configuration changes".

Best regards,
Victor

Hi,

we have request for simplifying SNMP trap configuration so there should be improvements in 3.x branch in July/August.

Best regards,
Victor

Yes, unmanaged state prevents SNMP transport creation. As possible workaround you can set node as managed from the script, create transport, and set it back to unmanaged. If you don't want any poll to start, you can disable polling before setting node to managed state.

Best regards,
Victor

This question was not related to UI - it was about using nxshell (Python based external scripting tool). You can get filtered information form event or alarm log in nxshell, but it won't be shown as UI view. You can do the same within event log or alarm log view by adding appropriate filters.

Best regards,
Victor

На какой платформе? На Линуксе вроде как давно сделан.

Сервер получает список доступных параметров во время configuration poll. Возможно он запрещен или еще не прошел.

Странно, по идее достаточно отключить network discovery polling чтобы ARP таблицы не опрашивались. Попробуйте включить дебаг уровня 5 и посмотреть в логе что пишется про этот узел.

Фикса наверное нет, надо разбираться с конкретной ситуацией. Есть инструкция по сборке консоли мавеном (из корня исходников):

Десктоп:

mvn -f src/java/netxms-eclipse/pom.xml -Dtycho.mode=maven org.eclipse.tycho:tycho-versions-plugin:update-pom -Pconsole
mvn -f src/java/netxms-eclipse/pom.xml clean package -Pconsole -Dtycho.disableP2Mirrors=true


Веб:

mvn -f webui/webapp/pom.xml -Dtycho.mode=maven org.eclipse.tycho:tycho-versions-plugin:update-pom -Pweb
mvn -f webui/webapp/pom.xml clean package -Pweb -Dtycho.disableP2Mirrors=true


Возможные значения инстансов можно получить через соответствующие агентские списки: Hardware.Batteries, Hardware.MemoryDevices, Hardware.Processors, Hardware.StorageDevices. В большинстве случаев это значение handle из SMBIOS таблиц.