[asterisk-users] console color

[Damon Estep]

I agree, but what is odd is that technically the same init script should
be used in both cases, since heartbeat initially starts asterisk with
the service start command, and service restart asterisk is used to
restart it.

 

perhaps safe_asterisk is restarting asterisk after the stop portion of
the restart and before the start portion? and perhaps when this happens
is when I get the "exited on signal 9", because the ports are already in
use as the result of the safe_asterisk recovery, so service asterisk
start (the other half of restart) fails?

 

any ideas on how to modify safe_asterisk to have the same console
parameters as asterisk?

 

From: asterisk-users-bounces at lists.digium.com
[mailto:asterisk-users-bounces at lists.digium.com] On Behalf Of Danny
Nicholas
Sent: Friday, September 18, 2009 8:36 AM
To: 'Asterisk Users Mailing List - Non-Commercial Discussion'
Subject: Re: [asterisk-users] console color

 

Just a wild guess, but your "service" probably runs two "flavors" of
asterisk.  Flavor 1 is /usr/sbin/asterisk  (executable) which provides a
console as you expect.  Flavor 2 is /usr/sbin/safe_asterisk
(shell-bash) which turns off the console color.

 

________________________________

From: asterisk-users-bounces at lists.digium.com
[mailto:asterisk-users-bounces at lists.digium.com] On Behalf Of Damon
Estep
Sent: Friday, September 18, 2009 9:28 AM
To: asterisk-users at lists.digium.com
Subject: [asterisk-users] console color

 

Hoping someone can help me understand what is happening here;

 

we start asterisk as a service at boot (actually, with heartbeat) on
CentOS using the asterisk init script installed with "make config"

upon reboot of the server (when the asterisk service is first started by
heartbeat) we get color in the console when we connect to it using
asterisk -r

after the execution of '#service asterisk restart' we no longer have
color in the console

 

this appear to be the case in all versions tested (1.2, 1.4, and 1,6)

 

additionally, when executing the restart of the service we get a message
that asterisk exited on signal 9, but I have not been able to find a
definition for signal 9. I assume this is normal because we force an
unconditional restart.

 

we do the restart periodically due to some processes that don't always
clean up after themselves, and the fact that a reload does not clean
them up either (zombie channels, zombie manager connections). these are
very heavily loaded servers, and the idea that a full restart should
never be needed has been proven inaccurate over several years of
experience :)

 

I do not think this is heartbeat related, but just in case, here are the
heartbeat details;

 

these are heartbeat version 1 clusters

the asterisk init script that is used is derived from "make config"

we chkconfig --add asterisk, then chkconfig asterisk off (heartbeat
starts it)

we then define the asterisk service as a heartbeat managed resource

about once a month we issue a "service asterisk restart" via a cron job,
and this is where we lose the color.

 

Thanks!

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.digium.com/pipermail/asterisk-users/attachments/20090918/c6af3ad9/attachment.htm