Just want to point out although there are many great fail over solutions
for asterisk, you should always have deadlock detection and scripts to
compensate as a deadlock will not trigger failover in many circumstances.
Jim Van Meggelen wrote:
Sangoma's TxTristate allows this to be automated.
If you are already using heartbeat, this is really simple.
Basically the steps are
1. edit /etc/wanpipeX.conf, and set
FE_TXTRISTATE = YES
2. configure heartbeat to run the shell command
wanpipemon -i w1g1 -c Ttxe
which turns on the transmitter on the T1 card of whichever system is
designated as the primary.
That's it.
It's totally automated, and takes less than 10 seconds to failover
from primary to secondary.
The secret to this is that the Sangoma drivers are loaded on both
systems, and so is Zaptel, but the transmitter on both the primary and
secondary card is turned off. When heartbeat decides which system is
going to be the primary, it enables the transmitter on that card, and
starts asterisk. In a failover scenario, heartbeat will do the same
thing on the secondary: Enable the transmitter and start asterisk.
Jim
Andre Courchesne wrote:
Hi Jim,
Thanks, damed, I'll have to talk to y boss so he pays me to attend
the TAUG meetings :-)
For the sync of the 2 asterisk box, that's not a problem, we
already have that and rely on a human to switch the PRI if ever there
is an issue.
But now we have a customer that requires an automated way.
----
Andre Courchesne
Concepteur Logiciel - Software Developer
[email protected]
PrivalODC Inc.
9955 ave Catania, local 145
Brossard, QC
J4Z 3V5
Web.: http://www.prival.ca
Tel.: (450) 761-9973 poste 635
1-866-761-9973
Fax.: (450) 761-9842
Ce message électronique ainsi que tous les documents annexés s’adressent
exclusivement à la personne ou à l’entité inscrit dans la rubrique
destinataire ; il peut contenir des renseignements de nature
confidentielle ou privilégiée aux termes des lois applicables. Nulle
autre personne ne doit y avoir accès. Si vous n’êtes pas le
destinataire convenu, nous vous avisons par la présente qu'il est
strictement interdit d'en divulguer le contenu, de le distribuer, le
copier ou l’ utiliser. Veuillez aviser l’expéditeur immédiatement par
retour de courrier électronique et supprimer ce message de votre
système. Toute diffusion ou reproduction de ce document ainsi que tout
mesure prise à l’égard de la présente est formellement interdite.
Merci de penser à l'environnement avant d'imprimer ce courriel.
Jim Van Meggelen wrote:
To me, that device represents a single point of failure, so I'm not
sure
what sort of redundancy that gets you. Also, you still sync between
your
two servers, so that device won't solve all your problems.
I'm going to do a talk for TAUG on how to use Sangoma's TxTristate mode
and heartbeat to create a hot-standby system, where you don't need any
external hardware.
It's just an overview, and heartbeat requires a fair bit of Linux
knowhow. But still, it's do-able, and the only single point of failure
then becomes the carrier equipment.
Jim
Andre Courchesne wrote:
Hi,
Anyone has experience with this device?
http://www.failsafevoip.com/index.php?main_page=product_info&products_id=1%22
Or anyone has recommendation for PRI failover switches ?
And no a human hand is not an option :-)
--
--
Jim Van Meggelen
[email protected]
http://www.oreillynet.com/pub/au/2177
"A child is the ultimate startup, and I have three.
This makes me rich."
Guy Kawasaki
--
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
