NetXMS Support Forum

And what is actual file size?

Hi,

try to run file transfer manually from server with nxupload command line tool - when will it stop and with what error? If possible, capture TCP session between server and agent used for upload and provide capture file.

Best regards,
Victor

Hi all!

We just published another patch release for version 3.8. Changes since previous patch release:

- Database manager can detect and delete orphaned data tables
- Event processing macro %A (alarm message) also works in rules following one that triggered alarm as well as when alarm is resolved or terminated by the rule
- Fixed agent crash on log parser rule with inverted match
- MIB compiler accepts object identifiers that starts with uppercase letter
- Implemented hash map to string conversion in NXSL
- Fixed save of node state
- Web API included into Web UI installer on Windows
- Fixed deployment issues with Web UI installer on Windows
- Fixed issues:
        NX-227 (Create "nxdownload" utility)
        NX-553 (CPU and Memory DCI support in Darwin)
        NX-1006 (Node with all status poll sources disabled creates a down alarm)
        NX-1057 (Ability to execute set of SQL commands on server start)
        NX-1473 (System Description for Agents that are also running SNMP should be Agent's description, not SNMP's)
        NX-1851 (nxdbmgr not checking for situation when DCI present in items, but not in object_properties)
        NX-1882 (Add option for negative mask in File.Count, File.FolderCount, and File.Size)
        NX-1947 (Server repeats SYS_IF_DOWN after SYS_NODE_UP even if interface was already down before SYS_NODE_DOWN)
        NX-1976 (Add interface names to automatically generated IP topology maps)
        NX-2043 (Delete "last agent contact" timestamp when isAgent capability switches off)

Best regards,
Victor

It's probably good idea to grant write access to whole \NetXMS\var - service may write other files there as well (most likely \NetXMS\var\images if you will add custom object images).

Best regards,
Victor

Hi!

Try to add the following option to netxmsd.conf:

DBCacheConfigurationTables = no

Best regards,
Victor

Повторил у себя. Похоже по умолчанию в агенте слишком маленький таймаут на выполнение внешних процессов. Добавил в конфиг

ExecTimeout = 30000

и рестарт сервисов заработал.

Скорее всего у текущего пользователя нет прав на шаблоны. Попробуйте зайти пользователем system, у него всегда есть все права, и проверьте права доступа в свойствах объекта Templates.

Hi,

the tricky part is what to consider uptime. If purely time when node was unreachable (like 100 - (intervals between SYS_NODE_DOWN and SYS_NODE_UP) / period length) then it is quite simple. In fact it can be done on current version with reporting. If uptime should include more conditions, like that all services are running on the node, then it became more complicated. This is what business services were meant for, but they are very hard to configure and definitely need rework.

Best regards,
Victor

Save the date – April 29, 2021 at 5pm CET (11am EST).

We will organize online Q&A session with NetXMS developers. Please contact @moshenska in our Telegram support channel or at [email protected] to sign up.

Settings are correct, but are server really listen on that port? You can check with command

netstat -aunp | grep 514

you should see netxmsd process listening on that port.

Best regards,
Victor

Please show screenshot of DCI configuratio and agent config file. Also, set agent debug level to 6, start agent, do force DCI poll on one of problematic DCIs, and show agent log file.

Best regards,
Victor

Hi,

when using agent table for instance discovery instance names are taken from instance column in the table, in case of System.Services it will be service name. So $1 in filter script is always instance, you have no access to actual table row. I suggest to use script as instance discovery method instead of agent table, and process table i that script instead. It could looks like following:


services = $node->readAgentTable("System.Services");
if (services == null)
   return null;

instances = %();
for(r : services->rows)
{
   name = r->get(0);
   startup = r->get(4);
   if (name like "3CX*" and startup == "Auto")
      instances->append(name);
}

return instances;


Best regards,
Victor

From nxmc you see syslog records received by NetXMS server process. Check that uinternal syslog receiver is enabled and listening on correct port, and that you forward syslog messages to correct port. As you can see messages from external nodes in /var/log/messages on NetXMS server I suspect that system syslog daemon receives them, not netxmsd process.

Best regards,
Victor

Hi,

it's hard to tell from provided information. How do you collect those metrics? It looks like something custom. If those are external parameters on agent, check that agent is up and running, and check agent log for possible errors.

Best regards,
Victor

Hi,

problem is definitely with pollers pool usage, rest looks fine. I would suggest gradually increasing size of pollers pool, probably start with 750/1500 from current 250/500.

Do you use hook scripts?

Do you have routers with huge routing tables? If yes, you could try to disable routing table poll for such devices.

If netxmsd is crashing, does system generates core file? If sould be really helpful for debugging the issue.

Best regards,
Victor