NetXMS Support Forum

Oops, looks like a bug. As a workaround, you can use legacy console to edit XML.

Best regards,
Victor

Did you restart agent after adding ping subagent? Also, can you check that you can get required value via nxget?
And maybe post here screenshot of your DCI config, this can help identifying the problem.

Best regards,
Victor

Hi!

You have to create correct log parser. See this section of the manual: http://wiki.netxms.org/wiki/UM:Log_Monitoring for parser syntax.

Best regards,
Victor

Hi!

Editor is not fully functional yet. You have to edit XML directly.

Best regards,
Victor

Sorry, my mistake - I mean "Parameter" field. "Name" was an old name for it. See attached screenshot for example. Description is a human-readable text to be displayed in UI, and "parameter" is an actual parameter name to be collected (it will be SNMP OID for SNMP parameters).

Best regards,
Victor

Это баг, который периодически всплывает в различных частях веб интерфейса - особенности переноса desktop клиента в веб. Мы их исправляем по мере обнаружения. Сейчас для не работаюсчих операций можно воспользоваться desktop клиентом.

Hi!

If I understand problem correctly - you want to see on node's performance tab DCIs created on interfaces, right? Then you have to find these DCIs in node's data collection configuration, go to properties, and check "show on performance tab" checkbox.

Best regards,
Victor

Добрый день!

Здесь есть два пути:

1. Разрешить серверу ходить на узлы по SNMP - судя по тому, что интерфейсы называются unknown, сервер не имеет доступа к этим хостам. Тогда сервер получит список интерфейсов и будет знать, что ето одна нода.

2. Создать интерфейсы руками - выбрать ноду, и потом Create interface из контекстного меню.

Не забудьте только вначале удалить "лишние" обьекты хостов.

Hi!

It's not so simple to use command's output in Windows as in UNIX shell. You can achieve this using FOR command:


for /F %x in ('nxget localhost Icmp.Ping(10.194.133.1)') do @nxpush -u admin -P ******** localhost Switch_1:Time=%x


However, there are easier way to achieve what you want using proxy DCI: create DCI on node Switch_1, set origin to "NetXMS Agent", enter Icmp.Ping(10.194.133.1) as name, and set proxy node to your NetXMS server node. This way system will show collected data for node Switch_1, but real data collection will occur from NetXMS server node. You will also need to add Switch_1 to list of trusted nodes in NetXMS server node's properties, or set server configuration variable "CheckTrustedNodes" to 0.

Best regards,
Victor

Hi!

Yes, support for SNMP tables is not finished yet. I have created a new issue in bugtracker so you can follow a progress: https://www.radensolutions.com/chiliproject/issues/198.

Best regards,
Victor

Normally, when alarm's repeat count increases (because new event was matched to existing alarm by alarm's key), alarm's state reset to "outstanding". To prevent this, you can use "sticky acknowledge". In that state, alarm will remain acknowledged event after new matching events.
This can be useful when you know that there will be new matching events, but it will not change the situation. For example, if you have network device which will send new SNMP trap every minute until problem solved, sticky acknowledge will help to eliminate unnecessary outstanding alarms.

Best regards,
Victor

Hi!

Thanks for pointing me on this! The housekeeper code is so wring here... One possible solution could be adding an index on event_log.event_source and replacing

SELECT event_source FROM event_log WHERE event_source=?

with

SELECT COUNT(event_source) FROM event_log WHERE event_source=?

and probably with

IF EXISTS (SELECT event_source FROM event_log WHERE event_source=50220) SELECT 1 ELSE SELECT 0

for MS SQL.

Another option is to remove this check altogether, as it is needed only for one purpose - prevent creation of new object with ID which was used before and still referenced in event log. This can happen, however, only if object(s) with highest IDs deleted, and server restarted before creating new object. It's probably better to just keep track of highest used ID elsewhere.

Best regards,
Victor

Collected DCI data stored in multiple tables - one table per node. These tables are called idata_nodeid, for example, idata_100.

You can also use Java API to access collected data from external application.

Best regards,
Victor

Hi!

You have to manually add network services you want to monitor. Right-click on node object, and select "create network service".

What kind of problem you have with transformation script?

Best regards,
Victor

Нет, не используем. А зачем он она нам может быть нужна (если я правильно понял, что вопрос о системах вроде MS BizTalk или FLink)?