Mike,
You should probably have a look at the tracker (http://track.sipfoundry.org). Not only does it track bugs but also, feature requests and, basically, provides a roadmap for the project. In the tracker you can vote for the features that you would like to see implemented as well as open a ticket for a new feature that you would like to see implemented. Collectively, we influence the priority of each feature or bug and whats included in the next version. I think the tracker will clear-up a lot of what is in the proposed plans for you. -Paul -----Original Message----- From: li...@grounded.net [mailto:li...@grounded.net] Sent: Sunday, August 30, 2009 1:21 PM To: sipx-users Subject: Re: [sipx-users] Where's the high availability? >The timestamps on your mail are 12 hours ahead of reality -- you >probably want to fix that. Had to rebuild this workstation yesterday and it somehow lost a day. Thanks for pointing that out. > but it's far > better than having the entire PBX stop functioning when the server loses > its disk drive. Don't get me wrong, I agree. I'm not giving up on sipx for reasons posted, I'm just curious about the future of it and more importantly, if there's anything more I can do in the meantime. > availability. What HA features do alternative software PBXs provide > that you would prefer in the next release, compared to the features that > are now proposed for the next release? I've not seen the proposed plans and being a new user, I'm not sure what I can add but there are some things I wonder about. Considering that sipx doesn't deal with RTP, it seems that it would not have the same problems that other PBX's would have, such as asterisk which does handle the media. What I'd like to see would be allowing other servers in the cluster to take over the function of management server should a management server become unavailable. Seems that the servers can see each other as it is so sounds like there's a head start on this already. And since DNS is really tied to services and not one individual server, that also seems to be a head start. I'd love to see things centralized, shared, over the network. The OS of course and other static things which don't much affect HA don't need to be shared. Web services, database, config files, all centralized. Like traditional clusters, the OS becomes the least important aspect of the cluster in that if a machine dies and needs to be rebuilt, it's a simple process to ISO a new machine up and running as just another node. Hardware is of course determined by budget and performance requirements. I use mostly fibre channel based network storage either directly attached to servers via HBA's or exported via filer heads. Network storage is very easy to do these days. There are plenty of fast NAS devices out there and those who don't have the budget could use a synchronization method instead or standard Ethernet based services such as iSCSI or NFS which is pretty darn fast and very reliable once set up right. Other than centralized data and a means of failing over, I'm not sure since I don't know sipx well enough yet but perhaps others can step in and offer some thoughts. Mike
_______________________________________________ sipx-users mailing list sipx-users@list.sipfoundry.org List Archive: http://list.sipfoundry.org/archive/sipx-users Unsubscribe: http://list.sipfoundry.org/mailman/listinfo/sipx-users sipXecs IP PBX -- http://www.sipfoundry.org/
