Use "any" for the IP of the agents. The IPs have to be unique (or a
CIDR range, and /32 is too small of a range).

On Fri, Oct 15, 2010 at 4:35 AM, tux3132 <[email protected]> wrote:
> I
>
> I have installed one agent on a (linux) host on a private network
> behind a firewall connected to the net with a static public IP
> address. This agent contact an Ossec server with also a static public
> IP address. The Ossec server is also behind a firewall on which the
> 1514/UDP port is redirected to the private address of this server.
>
> - On the server with manage-agent I give the public IP of the firewall
> behind which the agent reside.
> - On the client I have in ossec.conf:
>
>  <client>
>    <server-ip>public_IP_of_the_OSSEC_server</server-ip>
>  </client>
>
> When I import the key I have a good message.
>
> I run the agent and all is working fine.
>
> The problem arise when I add a new agent. On the server I have two
> agent with the same IP and the second agent can't connect to the
> server (the first one continue working fine).
> I have carefully read this doc : 
> http://www.ossec.net/doc/manual/agent/agent-dhcp-nat.html
> I have tried to add the agent on the server with a CID format of IP
> address : public_IP_of_the_agent/32 but no more job !
>
> Do you know how can I circumvent this problem because it mean that I
> need one Ossec server on each site ?
>
> Anticipated thanks.
>
> Best regards.
>
>

Reply via email to