NetXMS Support Forum

Hi!

Check that you have 127.0.0.1 in MasterServers.

Best regards,
Victor

Hi!

It's called "bind". You can select "Infrastructure Services" object or any container below, choose "Bind..." from context menu, and select node(s) you want to put into container.

Best regards,
Victor

Hi!

Internally server uses it's own agent running on same machine to monitor network services. You should check that NetXMS agent is running on NetXMS server's machine, and that it has port checker subagent loaded - there must be

SubAgent = portcheck.nsm

entry in nxagentd.conf. You don't have to install NetXMS agents on monitored nodes to check network services.

Best regards,
Victor

Hi!

We use Visual Studio 2005 for building NetXMS on Windows. I'll check and fix missing files. In the meantime you can use SVN snapshot to get these files: https://www.netxms.org/download/netxms-svn.tar.gz.

Best regards,
Victor

Hi!

Yes, it's a bug. Ill add correct handling of dormant operational status.

Best regards,
Victor

Hi!

How often do you see SNMP packets destined to devices with SNMP turned off? Can you show a screenshot of node properties and screenshot of SNMP request (or list SNMP OIDs requested by server)?

Best regards,
Victor

WMI пока работает только через агента установленного на хосте. Запланировано реализовать доступ к WMI без агента (https://www.radensolutions.com/chiliproject/issues/37), но пока не сделано.

Надо сделать 2 DCI:
1. prtMarkerSuppliesMaxCapacity - .1.3.6.1.2.1.43.11.1.1.8.1.1 - его можно собирать редко, поскольку max capacity вряд-ли меняется;
2. prtMarkerSuppliesLevel - 1.3.6.1.2.1.43.11.1.1.9.1.1 - собирать с необходимой частотой, и добавить transformation script, который будет пересчитывать абсолютное значение в проценты:

total = GetDCIValueByName($node, ".1.3.6.1.2.1.43.11.1.1.8.1.1");
return $1 * 100 / total;


Есть небольшой пример на английском по настройке параметра и отсылке сообшения: http://wiki.netxms.org/wiki/How_to_send_e-mail_notification_when_disk_space_is_low.
Вообще сначала надо определиться, как будут собираться данные: через агентов или через SNMP. С агентами все проще: ставится агент на хост, делается configuration poll, после чего кнопка "select" в диалоге конфигурации DCI будет давать список доступных параметров с агента. Для сбора данных с SNMP надо выбрать source "SNMP" в настройках DCI, после чего кнопка "select" откроет MIB browser, где можно выбрать нужный OID.

Hi!

Can you show how a chart of CPU usage on your NetXMS server? It seems quite strange to have very high CPU usage on that number of nodes. Regarding number of pollers, I would suggest to set it as following:

NumberOfConfigurationPollers = 20
NumberOfDatabaseWriters = 4 (or more if database writer queue still big)
NumberOfDataCollectors = 50
NumberOfRoutingTablePollers = 20
NumberOfStatusPollers = 50
NumberOfTopologyPollers = 20

High CPU usage may be caused by frequent polling of large routing tables. For example, if you have a router with full BGP table, it is recommended to turn off routing table polling for it, otherwise NetXMS server will try to fetch this huge table periodically.

Best regards,
Victor

You can send it to [email protected].

Best regards,
Victor

Hi!

As Android client is mostly read-only, you only need read access to the objects. Also you may need "acknowledge alarms" and "terminate alarms" rights if you need to manage alarms from mobile client. And running some types of object tools may require "Control" access right on appropriate nodes. You don't need any system-wide rights for mobile device users.

And other questions probably will go to Marco, as he is main author and maintainer of Android client.

Best regards,
Victor

Hi!

Did you place netxms-base_1.2.5_i386.deb along with netxms-agent_1.2.5_i386.deb?

Best regards,
Victor

Hi!

Try to wait for some time. Sometimes it took few topology polling cycles for server to discover all necessary information. You can also try to do topology poll on your switches manually (by selecting Poll -> Topology from context menu on switch object). If you still not able to get connection point information for node, check that it's really in switch forwarding database - right click on switch object, select Tools -> Info -> Switch forwarding database from switch object context menu, and check that node's MAC address is really there.

Best regards,
Victor

Скорее всего это из-за того, что устройства не поддерживают IP MIB, соответственно все интерфейсы без IP адреса, и при discovery сервер при проверке на наличие IP адреса в базе не находит его, и добавляет ноды еще раз. Я попробую добавить корректную обработку таких устройств.