Instead of further hijacking a specific question about call recording, I'll start a new thread... It's common to have more than one FS media server to handle calls. So then, you'll have them handling voicemail too. How do you ensure that the voicemail will be available for the user? Surely this is a common thing! Some options present themselves: 1) Ensure that each user always connects to the same server (this make maintenance impossible if the VM is only on one machine!) 2) Mount the VM directory to point to Amazon S3 (this doesn't sound ideal. Perhaps for long-term archival -- if you allow such a thing -- and seemingly the network connection could be a problem) 3) Some sort of NFS - but then this become a single point of failure? 4) Some sort of network drive that has multiple copies of the file, and it could even be local -- suggestions? Gluster? I've never used any. This would be transparent to FS. 5) couchdb, riak, (rethinkdb) (effectively) have multi-master replication/multiple sharding, but FS doesn't know how to access that. (In couchdb it's not a shard, it would be a full copy. The others configure a multiple of copies that are available) 5) Abaci suggested possibly record_post_process_exec_api or if it's not in VM that you can probably add a var in mod_voicemail.c to the file_path and use that var in the post_process api/app. Then, you trigger whatever "Copy" script to the NFS. 6) Jay Binks suggest http://code.google.com/p/lsyncd/ or using inotify directly, but it sounds like you need to establish a watch per folder, and simply - it's a hack. This should be something that can be triggered (if not fully handled) in FS itself. What are people currently doing to "share" voicemail? Can we work out some sort of recipe to easily implement #4 (or #5)? -Avi Marcus BestFone
On Fri, Dec 28, 2012 at 12:57 AM, Avi Marcus <avi@avimarcus.net> wrote:
3) Some sort of NFS - but then this become a single point of failure?
ODBC and NFS is a great way to go. Yes, they are each single points of failure, but each can be made redundant. That's where I would start before spending too much time tooling around. Best, Gabe
Ah, I missed it that part... the DB listing of the VM files needs to be shared across the machines, too. -Avi On Fri, Dec 28, 2012 at 10:13 AM, Gabriel Gunderson <gabe@gundy.org> wrote:
On Fri, Dec 28, 2012 at 12:57 AM, Avi Marcus <avi@avimarcus.net> wrote:
3) Some sort of NFS - but then this become a single point of failure?
ODBC and NFS is a great way to go. Yes, they are each single points of failure, but each can be made redundant.
That's where I would start before spending too much time tooling around.
Best, Gabe
_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com
FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com
Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com
FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
ODBC isn't necessarily a single point of failure - you can connect to a MySQL Cluster database or a multi-master circular replicated database for example. On 28 December 2012 08:13, Gabriel Gunderson <gabe@gundy.org> wrote:
On Fri, Dec 28, 2012 at 12:57 AM, Avi Marcus <avi@avimarcus.net> wrote:
3) Some sort of NFS - but then this become a single point of failure?
ODBC and NFS is a great way to go. Yes, they are each single points of failure, but each can be made redundant.
That's where I would start before spending too much time tooling around.
Best, Gabe
_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com
FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com
Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com
FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
As for NFS... using a cluster filesystem on top of DRBD is one way to avoid the filesystem becoming a SPOF. On 28 December 2012 14:20, Steven Ayre <steveayre@gmail.com> wrote:
ODBC isn't necessarily a single point of failure - you can connect to a MySQL Cluster database or a multi-master circular replicated database for example.
On 28 December 2012 08:13, Gabriel Gunderson <gabe@gundy.org> wrote:
On Fri, Dec 28, 2012 at 12:57 AM, Avi Marcus <avi@avimarcus.net> wrote:
3) Some sort of NFS - but then this become a single point of failure?
ODBC and NFS is a great way to go. Yes, they are each single points of failure, but each can be made redundant.
That's where I would start before spending too much time tooling around.
Best, Gabe
_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com
FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com
Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com
FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
Or even glusterfs, I've had goo results with that. On 28 December 2012 14:21, Steven Ayre <steveayre@gmail.com> wrote:
As for NFS... using a cluster filesystem on top of DRBD is one way to avoid the filesystem becoming a SPOF.
On 28 December 2012 14:20, Steven Ayre <steveayre@gmail.com> wrote:
ODBC isn't necessarily a single point of failure - you can connect to a MySQL Cluster database or a multi-master circular replicated database for example.
On 28 December 2012 08:13, Gabriel Gunderson <gabe@gundy.org> wrote:
On Fri, Dec 28, 2012 at 12:57 AM, Avi Marcus <avi@avimarcus.net> wrote:
3) Some sort of NFS - but then this become a single point of failure?
ODBC and NFS is a great way to go. Yes, they are each single points of failure, but each can be made redundant.
That's where I would start before spending too much time tooling around.
Best, Gabe
_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com
FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com
Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com
FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com
FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com
Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com
FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
-- *Andrew Cassidy BSc (Hons) MBCS SSCA* Managing Director *T <info@cassidywebservices.co.uk> *03300 100 960 *F<info@cassidywebservices.co.uk> *03300 100 961 *E <info@cassidywebservices.co.uk> *andrew@cassidywebservices.co.uk *W <info@cassidywebservices.co.uk> *www.cassidywebservices.co.uk
participants (4)
-
Andrew Cassidy
-
Avi Marcus
-
Gabriel Gunderson
-
Steven Ayre