[VoiceOps] Is call forwarding in Polycom phones fundamentally broken?
robert.dawson at mindshift.com
Wed Dec 22 15:46:27 EST 2010
>>> The one case where the Polycom might act funny is with shared lines.
SCA will definitely remove the call forwarding option from the phone completely.
>>> This doesn't make sense... When using the client side CF, the Polycoms send a 3xx message in response to the INVITE. There is no dialog set-up, and no media exchanged.
This is 100 % correct - the phone simply does a redirect. Looking for the redirection in an XSLog trace is usually how I find out that someone has call forwarding enabled when they swear they don't.
>>> double check that 'voIpProt.SIP.serverFeatureControl.dnd' and 'voIpProt.SIP.serverFeatureControl.cf'
This stuff is pretty well documented in the Polycom SIP Admin Guide but not really touched on in the Broadsoft partner guide for Polycom.
>>> In general, I really like Polycom phones, but may have no choice but to switch to cisco/aastra/etc
To be honest I have not found an endpoint that works better with the Broadworks platform than Polycom. Cisco endpoints running SIP are a total joke and the only time we use them is when customers want the Cisco name on display. I do have to say that the 7945/65 using the XML based configs are much better than the 7940/60 and the color displays are really impressive looking - 100% better than the color Polycoms. Of course there has been no interop done for them with Broadsoft and Cisco only officially supports SIP on the CCM platform for them. I managed to find a particular firmware version that actually has working NTP support while not breaking other features and cobbled together the config to get them working pretty well.
From: voiceops-bounces at voiceops.org [mailto:voiceops-bounces at voiceops.org] On Behalf Of Michael Bain
Sent: Tuesday, December 21, 2010 7:10 PM
Cc: voiceops at voiceops.org
Subject: Re: [VoiceOps] Is call forwarding in Polycom phones fundamentally broken?
On 21 December 2010 14:50, JD <jdupuy-list at socket.net> wrote:
> In client-side mode, the phone itself does the forwarding. If the
> phone is unreachable for any reason, the call forwarding never
> actually happens. That is because the call server doesn't know to
> forward the call and the phone isn't reachable to do the forwarding.
> By default, the call usually goes to VM, which is not what the user
> wanted. So, it works most of the time, but not predictably. So, it is broken by design.
This sounds like you want BroadWorks' "Call Forwarding Not Reachable"
service to cover this case.
> (Plus with the half-broken NAT support on the Polycoms, any forwarded
> calls needlessly burn two audio streams because the call is not likely
> to transfer off the handset. But, I'll refrain on Polycom NAT
> commentary for the
This doesn't make sense... When using the client side CF, the Polycoms send a 3xx message in response to the INVITE. There is no dialog set-up, and no media exchanged. BW will just resend the INVITE to the address in the Contact: header from the 3xx response. There is no further interaction with the forwarding phone.
> So, we go with server-side...
> In server-side mode, the phone refuses to show that the phone is
> forwarded anywhere on the screen. It is not uncommon for the user to
> "wonder why they aren't getting calls" if they forget to cancel the
> forwarding. So, it is also broken by design. (Albeit, not as badly as
> client-side, since at least it will always function.)
> This does not appear to be a bug, since it is well documented by Polycom.
The expected behavior of the feature is the behavior you want. The one case where the Polycom might act funny is with shared lines. With private lines, I'd double check that 'voIpProt.SIP.serverFeatureControl.dnd' and 'voIpProt.SIP.serverFeatureControl.cf' are enabled (and I'd recommend disabling 'voIpProt.SIP.serverFeatureControl.localProcessing.cf' and 'voIpProt.SIP.serverFeatureControl.localProcessing.dnd' as well. This way the phones will respect any override cases to the DND & CF
Enabling the serverFeatureControl features will cause the phones to start SUBSCRIBEing to the featureSync event package, so you can check your AS logs or a packet capture to confirm that it's being set up correctly.
If you've done all that, and you're still not seeing the behavior expected then I'd suggest opening a ticket with Polycom Support.
(I've not tested any of the above, and it's all off the top of my head so I may have missed something, but the essentials should be
VoiceOps mailing list
VoiceOps at voiceops.org
More information about the VoiceOps