NetXMS Support Forum

Hi,

it definitely looks like a bug. What is map zoom level when you experience this issue? Or what is approximate area covered?

Best regards,
Victor

I think ssh.nsm was never included into those packages (because it has dependency on libssh). We will discuss it internally, but likely solution is to build libssh and add it to the package together with ssh.nsm.

Best regards,
Victor

Does it appear if you relogin?

Best regards,
Victor

This is really strange. In your first post you mention interface FL-Signage with index 209 as having wrong IP address, but I don't see such interface in walk output. Could you please send me screenshot of overview page and interfaces page for that node? You can send it in PM if it contains sensitive information.

Best regards,
Victor

It could be due to timeout. Server will wait for answer from agent for time specified in server configuration parameter AgentCommandTimeout. If you are accessing remote system over slow link or command execution takes significant amount of time then executeSSHCommand may just timeout.

Best regards,
Victor

Hi,

yes, #1 is correct. Any way that get you working bcm2835 library is fine. Looks like download URL in that post is outdated.

#3 - you did rename nxmc-3.2.478.war to nxmc.war? If yes and still nothing shown check Tomcat log files or post them there.

Also, we have Raspbian repository for agent and server so you can just install them without building yourself.

Best regards,
Victor

Hi!

What exact errors you are getting? I see commands in your post but no actual output with error messages. Also, what version of Raspbian you are using?

Best regards,
Victor

Hi all!

NetXMS 3.2 patch release 5 is out. It fixes few important issues:

- Improved alarm browser performance when displaying big number of alarms
- Fixed memory leak in server
- Fixed issue NX-1821 (Scheduled Task script not accepting parameters)

Only server and user interface packages are affected.

Best regards,
Victor

А какая версия сервера? Но expandString должен работать, поскольку иначе была бы ошибка в той строке.
Попробуйте еще в trace добавить результат expandString, и потом сравните в аларм логе что ключ действительно совпадает с алармом.

Похоже что аларм не находится - в этом и проблема. Покажите пожалуйста как сейчас выглядит правило.

Hi all!

NetXMS patch release 4 (version 3.2.462) is published. This patch only affects server component. It fixes single issue with thread pool manager that can cause start of excessive number of threads in thread pools. All other components can stay on version 3.2.451.

Best regards,
Victor

Hi,

could you please provide output of SNMP walk for the following OIDs for this device:

.1.3.6.1.2.1.4.20.1
.1.3.6.1.2.1.4.34.1
.1.3.6.1.2.1.4.32.1
.1.3.6.1.2.1.2.2.1

Best regards,
Victor

Ну да, я не подумал что аларм сбросится в outstanding на повторном событии. Тут два варианта решения:
1. Сделать такой-же фильтр для генерации алармов - т.е. событие будет просто игнорироваться если аларм уже есть. Тогда можно отсылку нотификации и генерацию аларма объединить в одно правило. Единственный недостаток - счетчик повторов у аларма не будет увеличиваться.
2. Использовать "sticky acknowledge" вместо "acknowledge" для аларма - sticky состояние не будет сбрасываться обратно в outstanding.

В аттаче пример правил. Первое правило отсылает сообщение если аларма с заданным ключем нет или он в состоянии "outstanding". Второе правило создает аларм с ключом (или обновляет существующий если аларм с таким клячем уже был).

One possibility is to change interface naming in netxms. You can do that either in configuration poll hook or by using interface name templates. If interface object names will match then topology builder should puck them up correctly. You can use macro that calls a script and replace ETHERNET PORT n with just PORT n.

Best regards,
Victor