NetXMS Support Forum

Hi!

From the log I can see that agent successfully find new record in log and tries to send it to server:

[17-May-2013 23:14:08.683]   matched
[17-May-2013 23:14:08.683] SendTrap(): event_code=100008, event_name=(null), num
_args=1, arg[0]="Recibido comando de inicio de IIS del usuario HPCDSES\administr
ador. El dato registrado es el c¾digo de estado. " arg[1]="(null)" arg[2]="(null
)"

What I didn't see in the log is communications with the server. Please check that server can communicate with agent. You can run configuration poll and see if agent is detected. If not, check network connectivity and firewalls between server and agent.

Best regards,
Victor

Судя по логу, configure вместо C++ компилятора (g++) нашел некий gpp. Проверьте, что установлен пакет g++, и удалите gpp, если он реально не используется.

В новой консоли тоже можно делать автоматический логин. Вот документ с описанием параметров командной строки: http://wiki.netxms.org/wiki/Autologin_for_Management_Console.

Hi!

Did you try to configure any custom attributes for some object?

Best regards,
Victor

Hi!

Parser looks correct. It should generate an event for each critical event in Windows system event log. I suggest to check the following:

1. Agent config - you should have logwatch subagent, and you should have parser definition set in LOGWATCH section.
2. Check event log on NetXMS server - do you see events coming?
3. Check that there really was new critical events.
4. Run agent in debug mode and see if there are any errors/warnings related to log monitoring.

Also, don't be confused by syslog view in NetXMS - agent send events to server, not syslog records - and so you should see them in event log in NetXMS.

Best regards,
Victor

Hi!

Looks like there was some remnants from previous version. I encounter similar situation on Tomcat - you have to re-deploy application while Tomcat is running. If you stop Tomcat, copy over new war file, and start Tomcat, some packages from old version can remain in temporary directories.

Best regards,
Victor

Попробуйте запустить агент с полным дебагом (ключ -D9). И пришлите пожалуйста конфиг агента.

С тех пор сообщения менялись, я лучше соберу новый архив. Был человек, который начал перевод на русский, но от него давно ничего не слышно...

Hi!

Can you check please what driver was selected for these switches? Cisco has very specific way of providing MAC address table for VLANs other then VLAN 1, so this may be the case. Catalyst driver supposed to handle this, but maybe there are bugs in the driver, or correct driver was not selected at all.

Best regards,
Victor

Hi!

We know that there are installations with more then 3000 nodes. Potentially system can handle hundreds thousands nodes - actual capacity mostly depends on how often you collect data, how much data you collect, and your network bandwidth.

Best regards,
Victor

Hi!

Can you confirm that problem on 1.2.7 as well? There was some fixes in node polling. This bug is so strange...

Best regards,
Victor

Hi!

Script looks correct. But are you sure that .1.3.6.1.2.1.25.1.5.0 is writeable, and that community string set in NetXMS allows read/write access?

Best regards,
Victor

Hi!

You can turn on global event storm detection by setting server's configuration variable EnableEventStormDetection to 1 (you'll have to restart server). You can set number of events to trigger event storm condition in EventStormEventsPerSecond parameter, and for how many seconds number of events mus be above threshold in EventStormDuration parameter.
Another option is to implement counter using situations, and don't send email if certain threshold reached.

Best regards,
Victor

Hi!

Can you please attach your config.log?

Best regards,
Victor

Все настройки сервера хранятся в базе, так что достаточно перенести базу, проинсталлировать NetXMS сервер на другой машине, и подключить к скопированной базе. Базу можно перенести через nxdbmgr export / nxdbmgr import, или средствами самой базы. Если база SQLite, то достаточно просто скопировать файл базы.