I haven't played with this at all, but I don't think the agent will
make a connection to the backup manager until it notices that the
first manager isn't there anymore.

On Tue, Jan 25, 2011 at 9:06 AM, ItsMikeE <[email protected]> wrote:
> Following on from the discussions in this topic I have attempted to set up a
> high availability solution.
>
> All servers are running RHEL 5
> I had a working ossec master with /var/ossec on a separate filesystem.
> For the moment RIDS is still enabled.
>
> On a 2nd server I installed the ossec master software, then removed
> everything under /var/ossec, and stopped ossec processes from starting
> automatically on reboot.
> A mirror was created for the filesystem containing /var/ossec between the
> master and the 2nd master.
> I stopped ossec on primary master, made sure the mirrors were up-to-date,
> broke the mirrors and mounted the mirror on the new master.
> On one of my agents I amended ossec.conf to have two IP addresses
> <ossec_config>
>   <client>
>     <server-ip>original_master_IP</server-ip>
>     <server-ip>new_master_IP</server-ip>
>   </client>
> </ossec_config>
>
> I then restarted ossec on the agent, and started ossec on the new master.
>
> I have not been able to establish any communications between the agent and
> the new master.
> The UDP 1514 port is open (checked using nmap).
>
> I tried re-starting the agent from the new master
>
> ./agent_control -R 009
> 2011/01/25 14:02:17 agent_control(1210): ERROR: Queue '/queue/alerts/ar' not
> accessible: 'Connection refused'.
> 2011/01/25 14:02:17 agent_control(1301): ERROR: Unable to connect to active
> response queue.
>
> ** Unable to connect to remoted.
>
> Can anyone suggest where the problem is?
>

Reply via email to