Welcome to OCS Inventory NG community support, where you can ask questions and receive answers from other members of the community.

Please ask questions only in English or French.

Release 2.12.3 available

The official documentation can be found on https://wiki.ocsinventory-ng.org. Read it before asking your question.

Server with multiple interface doesn't discover via SNMP [closed]

I have an OCS management server mounted in a CentOs 6.5 The version of the OCS management server  is 2.1.2.
The agent version for Unix mounted in other CentOs 6.5 server (2.6.32-504.3.3.el6.x86_64)is OCS-NG_unified_unix_agent_v2.1rc1
After installing the agent on the server, automatically it appears on the web and it collects all the information correctly.
In that server, I've configured several network interfaces so that I can scan and discover other devices from other networks.
I've configured the server so that it can works with the scan feature and the log doesn't show any errors but the problem is that although the server has more than one interface, it only scans for the default interface.
The default IP of the server is 10.70.7.171/24
The other interface that I've added is: 10.71.137.10/24
On the management server, I've configured the SNMP comunity v2c (snmpprocs) too and the properties of the server Network scan are these.
Network scans
IPDISCOVER --> Revert to standard state
SNMP_SWITCH --> Default
SNMP_NETWOrK --> 10.71.137.0,10.70.7.0

SSL is configured and working properly on the server. In fact it discovers devices via SNMP but not from the IP range I've wanted.
Could anybody tell me if I've done something wrong or if it's impossible to discover devices in the way I want?
closed with the note: Solved
in OCS Inventory NG agent for Unix by (610 points)
closed by

1 Answer

0 votes
monica,

You are going to need to policy route each subnet you are wanting to scan out the proper interface.  I found an online turorial for you from Darian Kindlund's Blog that pretty much covers all the necessary steps you are going to need to implement to properly be able to retrieve SNMP, PING, TRACEROUTE, etc via different subnets/interfaces.

https://kindlund.wordpress.com/2007/11/19/configuring-multiple-default-routes-in-linux/

Good luck,
by (140 points)

The problem is not routing because if I do a ping to any device on any network on the command line, the result is OK.

The problem is that I don't know how to specify to the OCS application to scan all networks the server has and not for only one interface. In tahat way the application works perfectly.
Is there any way to do this I am saying in OCS?

 
Powered by Question2Answer
...