NetXMS Support Forum

Hi,

yes, it is planned for one of next releases.

Best regards,
Victor

Hi,

currently it's not possible. Feel free to register feature request for that, it should not be hard to implement.

Best regards,
Victor

It should be nxmc.ini, not eclipse.ini.

Best regards,
Victor

Hi,

you can put interface names in double quotes (it will work for instance discovery as well), like this:

Net.Interface.BytesIn("{instance}")

which will be expanded into

Net.Interface.BytesIn("RAS (Dial In)")

and should work correctly.

Best regards,
Victor

Hi,

CATALYST-GENERIC driver uses CISCO-STACK-MIB to map interfaces to physical ports. Unfortunately this MIB is not supported by 3650. It seems that we have to use ENTITY-MIB to find physical ports. Could you please provide result of SNMP walk on OID .1.3.6.1.2.1.47.1 on your switch (or even better switch stack)?

Best regards,
Victor

Hi,

server considers that first device found in certain subnet to be configured correctly (because usually it's a router). Sometimes it may encounter incorrectly configured device first. You can check all devices within subnet manually using "Interrfaces" tab, or you can create NXSL script that will walk nodes within subnet and report those with /8 mask on interface.

Best regards,
Victor

Hi,

actually NetXMS server should handle VRRP MACs differently, as they will not be unique. Zoning will not help because you may have multiple VRRP interfaces within same zone. As for the topology - usually routers also provide information about real interfaces with unique MACs and VRRP MAC address should be shown only for VRRP interface, so router could be placed into correct network using it's physical interfaces. If VRRP MACs will be excluded routers should be placed correctly in L2 topology.
I've registered issue for that: https://dev.raden.solutions/issues/1196

Best regards,
Victor

Hi,

map changes should be saved automatically as soon as you made them.

Best regards,
Victor

Hi all!

NetXMS version 2.0.3 is out. This release mostly aimed at increasing system stability and bug fixing. Full change log is following:

- Additional parameters in Oracle subagent for redo logs and ASM monitoring
- NXSL: variable $errormsg holding error description set in catch block
- New NXSL function SplitString
- NXSL: new methods setExpectedState and setExcludeFromTopology in Interface class
- Default interface expected state can be configured
- GPS subagent (provides location data from NMEA compatible GPS receivers)
- Server can update node location using GPS subagent
- Fixed error while event alarm linking multiple times
- Fixed server hang on shutdown
- NetXMS processes uses LC_CTYPE from environment for terminal I/O
- Fixed bug in saving service checks
- Fixed server crash caused by heap corruption in MS SQL and DB/2 database drivers
- Fixed server deadlock caused by simultaneous use of instance discovery and transformation scripts
- Agent parameters for monitoring System V message queues
- New action User.ChangePassword in WINNT subagent
- Text2reach SMS driver improved
- Port check subagent can be configured to return negative value as response time in case of error
- Management console:
        - Fixed grid issues in network maps
        - Fixed object selection issues in network maps
        - Added "hide links" option on network maps
        - Added support for MAC addresses in format xxx.xxx.xxx.xxx
- Fixed issues: #950, #962, #988, #1048, #1102, #1124, #1131, #1135, #1147, #1149, #1154, #1163, #1166, #1170, #1174

Best regards,
Victor

Hi,

currently it's not implemented. For 2.0.3 I've added new method for interface object - setExcludeFromTopology.

Best regards,
Victor

Hi,

can you share screenshot of "Interfaces" tab for the switch shown under 10.7 subnet?

Best regards,
Victor

I mean that I just added it You have to wait for 2.0.3 release or build agent from git repository.

Best regards,
Victor

Hi,

dot (.) is a string concatenation operation, so line

data=cfunctions.getToken(data);

concatenates value of variable cfunctions (likely null) with returned array, which resulted in a null converted to string concatenated with array converted to string.

Best regards,
Victor

Hi,

I've added configuration parameter NegativeResponseTimeOnError for portcheck subagent to return negative value instead of data collection error in case of service communication error. Config example:


MasterServers = 127.0.0.1, ::1, 10.5.0.0/24
LogFile = /opt/netxms/log/nxagentd.log
SubAgent = portcheck.nsm

[PORTCHECK]
NegativeResponseTimeOnError = true


Best regards,
Victor

Hi,

if you are running server scripts (NXSL), you can use PostEvent function: https://wiki.netxms.org/wiki/NXSL:PostEvent

Another possible option is to set node's custom attribute in reaction for SYS_NODE_DOWN and create status poll hook script, which will check if node is down for certain amount of time and use PostEvent function to generate another event.

Best regards,
Victor