NetXMS Support Forum

Hi,

configuration on agent side seems to be ok, and SendTrap record in log confirms that event is being generated. Check that you run agent with debug level 6 or higher and try to find record similar to "sending message CMD_TRAP" after SendTrap() record. If you'll find it, it mean that message was actually sent to server and problem is on server side. If not, problem is on agent side.

Best regards,
Victor

Hi,

first of all, I would recommend to change events for threshold to SYS_TABLE_THRESHOLD_ACTIVATED and SYS_TABLE_THRESHOLD_REARMED - you will get much more meaningful messages including service name. Other than that configuration seems fine (you can actually use == instead of like when you are not using metacharacters, but it should not make any difference in result).

Best regards,
Victor

I would recommend 8-16GB RAM and 4 CPU cores. For this number of nodes you'll have instant flow of INSERTs in database, so you may want to give most attention to I/O and MySQL tuning.

Best regards,
Victor

Этот параметр появился в 2.0-M1.

Hi,

you can force driver selection for a node by setting node's custom attribute snmp.driver to driver name. Try to use Catalyst driver and if it will work I'll correct supported device detection in the driver.

Best regards,
Victor

Hi,

do you mean tooltips on network map?

Best regards,
Victor

Just few clarifications:

1. load will not exceed size, bt it can exceed max size, as thread pool will not grow beyound max size.

2. Note that parameter name for load monitoring is "LoadAverage", not "Load", as it represents moving average of request queue for period (same as UNIX/Linux load average), while "Load" represents current thread pool load as % of threads processing request to current size of thread pool.

Also, I've updated internal DCI selector in UI - now it shows correct set of parameters.

Best regards,
Victor

I'll implemented it in 2.0-RC1 - change should be quite easy.

Best regards,
Victor

Hi,

seems to be a bug in NetXMS. Could you please show how interfaces on your nodes looks like (Interfaces tab in node details)?

Best regards,
Victor

Hi,

that's likely a problem. Currently NetXMS can only build topology if both peers provide LLDP information via SNMP. In this particular case ap1 reports that it identify itself using MAC address (.1.0.8802.1.1.2.1.4.1.1.4.0.57.2 = 4), but because it does not provide own LLDP ID via SNMP NetXMS cannot find node with this LLDP ID. I can add a fallback search using remote system name (.1.0.8802.1.1.2.1.4.1.1.9) - it may cause inconsistent results though if for example few nodes have identical system names.

Best regards,
Victor

Hi,

seems to be a bug in dashboard implementation. We will check.

Best regards,
Victor

Hi,

actually, LLDP is quite flexible in what is used to identify node - every peer sends it's identification along with ID type. It can be MAC address, host name, interface name, etc.). Can you please share results of SNMP walk on OID .1.0.8802.1.1.2.1.4.1.1 on switch1 as well as .1.0.8802.1.1.2.1.3 for ap1?

Best regards,
Victor

Hi,

it seems that at some point you forgot to specify prefix to configure and it installs binaries and libraries into default location (/usr/local). You can safely delete all netxms libraries from /usr/local/lib - then 2.0-M5 should work without problems.

Best regards,
Victor

Hi,

did you install it from sources or from binary package?

Best regards,
Victor

Hi,

I recommend to try 2.0-M5 - some memory leaks was fixed in that release.

Best regards,
Victor