NetXMS Support Forum

Hi!

I mean that in passive mode NetXMS uses data from known devices to find new devices. For example, you have one device in NetXMS - router with IP address 10.0.0.1 accessible with SNMP. During discovery poll, server reads ARP cache and routing table from that router. For example, it founds two entries in router's ARP cache - 10.0.0.7 and 10.0.0.44. NetXMS then adds new devices with addresses 10.0.0.7 and 10.0.0.44 (if filter does not prevent it), and do discovery poll on them if possible - and that process repeats for each found device. You need active discovery only if you have devices which may not communicating with already known devices, so they will not appear in known devices' ARP cache.

And yes, pinging all IP addresses on subnet can take very long time, especially if you have many unused addresses, because server will wait for possible reply for each address.

Best regards,
Victor

Hi!

Could you please provide exact command which does not work?

Best regards,
Victor

You should create two DCIs, and transform one of it using value from another. For example, define DCI for interface speed (.1.3.6.1.2.1.2.2.1.5.1); leave it as is. Then, define DCI for inOctets (.1.3.6.1.2.1.2.2.1.10.1), and set transformation script to


sub main()
{
   ifSpeed = GetDCIValue($node, FindDCIByName($node, ".1.3.6.1.2.1.2.2.1.5.1"));
   return ($1 * 8 / ifSpeed) * 100;
}


See also this topic: https://www.netxms.org/forum/index.php/topic,289.0.html

Best regards,
Victor

Hello!

In passive mode, NetXMS server only collects information about network devices by reading ARP caches and routing tables on already known devices. In active mode, server additionally pings every address in given subnets and uses responding addresses as well.

Network discovery can take many hours or even days, it depends on network size. Also, don't forget that NetXMS server should have SNMP access to as much network devices as possible to be able to obtain required information.

You can discover all reachable devices and networks, just make sure that your devices and firewalls allows SNMP requests from NetXMS server.

Best regards,
Victor

Test button will work only for nodes (not templates) - because test is done by sending script to server and running it in context of specific node.
For testing scripts where $node variable is not used you could also use nxscript command line tool. If using nxscript, first command line argument will be passed to script as $1, second as $2, and so on.

Best regards,
Victor

Ja poprobuju u sebja povtorit' etu situaciju. Kak workaround, poprobujte vibrat' "use existing database" - esli baza uze sozdana i prava na nee est', to vse dolzno srabotat'.

What exact error you got? Is it entire script or just a fragment? If it is entire script, you should write it as

sub main()
{
  if ($1 == null)
     return 0;
  else
     return ($1 * 2);
}

or rewrite as

($1 == null) ? 0 : $1 * 2

Best regards,
Victor

Hi!

Could you please try version 1.0.0? I've fixed some bugs in interface poller, should help.

Best regards,
Victor

Hi!

In this case, 3 is number of retries, and ping timeout is 1500 milliseconds. Unfortunately, this value is hard-coded (file src/server/core/interface.cpp line 329).

Best regards,
Victor

Hi!

This bug is inside console application - it is not related to server. I also have console hang from time to time, but still unable to find the source of the problem.

About logarithmic option: I have added functions log, log10, and exp to NXSL - they will be available in next release. I will also add logarithmic scale option to graph window in console.

Best regards,
Victor

Hi all!

Version 1.0.0 is out. It fixes some bugs found in 1.0.0-rc2, and contains minor improvements in server and Windows management console. Also, DRBD monitoring code in Windows subagent was rewritten to use proc file system and now should be compatible with modern DRBD versions.

Full list of changes since 0.2.31:

- System configuration (events, SNMP traps, templates, event processing rules) can be exported and imported in XML format - this solves various problems with old NXMP files
- NXSL improvements:
        - New functions: GetDCIObject, GetInterfaceName, FindNodeObject, trace
        - Added built-in constants TRUE and FALSE
- Added support for encrypted DB passwords in netxmsd.conf
- Improved SNMPv3 support
- Windows management console: implemented object search by comments
- Linux subagent: DRBD monitoring rewritten
- Fixed issues: #62, #278, #283, #290, #292

Best regards,
Victor

Spasibo, ispravil.

Hello!

I assume you have Windows workstations. There are no built-in mechanisms for initial agent deployment (but there are centralized upgrade feature). For initial deployment, you should use Active Directory functionality or some third party product for software distribution. To assist in this task, agent installer can run in unattended mode. Agent installer has various command line options for unattended installation. These options are following:

/SILENT Don't show installation wizard, only a progress bar;
/VERYSILENT Don't show anything;
/SUPPRESSMSGBOXES Don't ask user anything;
/DIR=path Set installation directory (default is C:\NetXMS);
/SERVER=address Set address of master server (will be written to generated agent's configuration file as value for MasterServers parameter);
/SUBAGENT=name Enable subagent name;
/NOSUBAGENT=name Disable subagent name.

Valid subagent names are ECS, LOGWATCH, PING, PORTCHECK, WINPERF, and UPS. By default WINPERF is enabled and all others are disabled.
For example, to install agent silently (without any messages or progress bars), use Windows Performance and PING subagents, and set NetXMS management server address to 10.0.0.1, run the following command:

nxagent-1.0.0-rc2.exe /VERYSILENT /SUPPRESSMSGBOXES /SERVER=10.0.0.1 /SUBAGENT=PING

Best regards,
Victor

Hi!

For System.ServiceState parameter, you should specify service name - don't be confused with display name you see in "Services" in MMC. Service name can be seen in service properties. System.ServiceState returns service status code, which can be one of the following:

0 - service running;
1 - service paused;
2 - service starting (start pending);
3 - service pausing (pause pending);
4 - service starting after pause (continue pending);
5 - service stopping (stop pending);
6 - service stopped;
255 - unable to get current service state.

So in threshold you have to check if status code not equal 0 and if it is, generate appropriate event. The following discussion could also be useful:
https://www.netxms.org/forum/index.php/topic,166.0.html

Best regards,
Victor

There are no limitations on client IP addresses. What exact error you get when connecting from remote PC?

Best regards,
Victor