On Fri, May 25, 2012 at 7:23 AM, <[email protected]> wrote: > 2) Use UC DNS for the servers and the company DNS for the clients, this > means putting the > necessary records in the company DNS. All records are available on the UC > servers as well. > This is the safest I think because the client records are available > everywhere. No changes on DHCP. > 3) Use company DNS for all, meaning entering all records in the company DNS. > This requires special skills and time, but would work OK.
I don't see any advantage of #3 over #2. If you disable/or enable a single service in HA, it affects DNS now more than ever, but only for internal connections between UC components so why open a ticket with IT to make DNS change when the change does not affect any endpoints. What we need it a clear list of endpoint related records you should copy into your company DNS server. It may help not to think of the DNS server embedded in UC as a "DNS Server" but instead as an "name resolution system that happens to follow the DNS specification" This is a good discussion so thanks. Fortunately the distinction between #2 and #3 is really just a best practices thing and nothing sipxecs has to do or not do. _______________________________________________ sipx-users mailing list [email protected] List Archive: http://list.sipfoundry.org/archive/sipx-users/
