I appreciate the quick response.  Well I guess I can ask about the real
problem, which is with prelude support:

When starting ossec, the analysisd daemon is started and it is supposed to
create the queue/ossec/queue which is where the other daemons
(log-collector, syscheck) send their events.  However, before creating that
queue, it tries to initialize prelude and this can cause problems.
Networking problems can cause prelude_start() to take over a minute to
return, so meanwhile the other daemons have been started and are trying to
connect to the queue (queue/ossec/queue) which doesn't exist because the
analysisd daemon hasn't created it yet because the prelude_start() function
hasn't returned yet.

Has anybody else run into this? Is this a way around this?  I have a very
kludgy solution and I want to know if there are any better ways to start
ossec successfully even if prelude takes a while to timeout.

Thanks,

Kevin

On Mon, Nov 28, 2011 at 8:22 PM, dan (ddp) <[email protected]> wrote:

> You'll have to do some surgery on the code, possibly a lot. It'd
> probably be easier to solve the problems you have than adding more.
>
> On Mon, Nov 28, 2011 at 8:04 PM, kevin sullivan
> <[email protected]> wrote:
> > I have been having a couple issues with running ossec locally with
> prelude
> > support enabled and one of the solutions I think would work is if I could
> > run ossec in a non-chrooted environment.  Is there information on how to
> run
> > ossec without chroot-ing, or does ossec need to be run in a chrooted
> > environment for reasons I don't know about?
> >
> > Thank you,
> >
> > Kevin
> >
>

Reply via email to