NetXMS Support Forum

Переделал на poll - изменения внесены как в ветку develop, так и в stable-2.0. Можно пробовать.

У субагента ограничение на размер буфера - 10МБ. Но вряд-ли документ больше 10МБ? Ече можно включить сохранение полученых страниц, добавив параметр FailedDirectory в секцию PortCheck в конфиге агента - тогда на каждый неудачный запрос будет сохранен html, можно будет проверить что в нем реально находилось.

Посмотрел дамп - единственно что вижу - много вызовов kevent. Похоже мы как-то неправильно его используем. Сегодня-завтра переделаю, можно будет попробовать.

Hi,

so weird, it was supposed to be there... Seems to be backport issue. I've fixed it in current 2.0 branch, it will be part of 2.0.8 release if it will happen.

Best regards,
Victor

Try to set server debug level to at least 5 and check what is going on during login.

Best regards,
Victor

Hi,

you can try to reset admin password first (with netxmsd stopped) using nxdbmgr resetadmin command, and then try to login with admin/netxms.

Best regards,
Victor

Registered as feature request: https://dev.raden.solutions/issues/1354

Hi,

yes, it can be done. Could be fairly easy on node level - is just another poll, like status or configuration. On a subnet level will take a bit more work, although not much.

Best regards,
Victor

Hi,

it seems that agent is not responding to the server on that machine. Please check what IP addresses are listed as MasterServers in nxagentd.conf. If you have multiple interfaces on management server make sure you list them all. Also check what is set as "primary host name" in communication properties for that node.

Best regards,
Victor

Hi,

it shows if you have to restart netxmsd to put changes in particular configuration parameter into effect.

Best regards,
Victor

Hi,

yes, if script returns false instance will be deleted. Currently there is no way to stop instance discovery by aborting filter script.

There is feature request for adding "grace period" for instances being deleted: https://dev.raden.solutions/issues/976. I think implementing it should solve your problem.

Best regards,
Victor

Hi,

currently vmgr subagent is available only for Linux. It is build using libvirt which is not properly ported to Windows (some older versions was, but not current version). We will work on Windows port as well, but currently you have to use Linux machine.

Best regards,
Victor

Hi,

yes, for /32 subnets are not created. Those are nodes without agent or SNMP, right? Do you have some subnets that are close to addresses discovered? If yes, what subnets?

Best regards,
Victor

Hi,

do you have source node set for this DCI? Part of agent database that you show - is it from correct node or from node where this DCI should not be?

Best regards,
Victor

Hi,

what you mean by "cannot generate L2 map"? If you create new map with type "layer 2" and seed node set to your central router, what you will get on that map?

Best regards,
Victor