[tac_plus] TACACS+ Error socket issue.

Alan McKinnon alan.mckinnon at gmail.com
Thu Mar 29 22:34:26 UTC 2012


On Thu, 29 Mar 2012 23:36:51 +0200
Peter Mraz <trubela at gmail.com> wrote:

> Hi,
> 
> Do you know how could we verify If it died?
> Can happen that we will would like to kill more tac_plus at the same
> time(each with specific name like R1_tac_plus.conf or
> R2_tac_plus.conf, each running on different port.)

cat the pid file in

/var/run/tac_plus.pid.<portnumber>

and check if a process with that pid is running.

You are starting tac_plus correctly and it is very rare for the daemon
to not die properly (or indeed to not respond properly to SIGs).

The best way to start|stop tac_plus  is to use the Red Hat init.d
system. You have compiled tac_plus yourself so you will have to either
write your own /etc/init.d/tac_plus file or get one somewhere, then use
"service tac_plus start|stop" as normal.

Red Hat should provide rpms for tac_plus but most likely they will be
way out of date. I suggest installing one of these rpms on a spare
machine anyway, copy the init.d script for use elsewhere the remove the
rpm package.



> 
> Cheers.
> 
> 
> On 29 March 2012 19:55, heasley <heas at shrubbery.net> wrote:
> > Thu, Mar 29, 2012 at 07:37:34PM +0200, ivo:
> >> kill -9 <pid>
> >> /usr/local/bin/tac_plus -C /etc/tacacs+/R1_tac_plus.conf -l
> >> /var/log/tacacs/R1_tac_plus.log -d 16 -p 5000
> >>
> >> I recieve error into R1_tac_plus.log:
> >> Version F4.0.4.19 Initialized 1
> >> tac_plus server F4.0.4.19 starting
> >> Backgrounded
> >> Error get_socket: bind 5000 Address already in use
> >
> > make sure that it has actually died.
> 
> 
> 



-- 
Alan McKinnnon
alan.mckinnon at gmail.com



More information about the tac_plus mailing list