|
Home > Archive > Voice over IP Cisco > November 2005 > Route patterns, translation patterns
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Route patterns, translation patterns
|
|
| Candace Holman 2005-11-15, 5:46 pm |
| Can anyone send me a cisco reference for the order in which
route/translation patterns are followed?
For a legacy SMDI integration we need to change calling party number on
calls, but only after they have forwarded to voice mail, and only on
some of the calls. If anyone has done something like this, can you
explain how it was done, and which version CM you needed to run?
Thanks,
Candace
| |
| Wes Sisk 2005-11-15, 8:45 pm |
| Use the voicemail profile to mod the originalcalledparty /
originalcalledpartyvoicemailbox.
/Wes
Candace Holman wrote:
>Can anyone send me a cisco reference for the order in which
>route/translation patterns are followed?
>
>For a legacy SMDI integration we need to change calling party number on
>calls, but only after they have forwarded to voice mail, and only on
>some of the calls. If anyone has done something like this, can you
>explain how it was done, and which version CM you needed to run?
>
>Thanks,
>Candace
>
>
> ________________________________________
_______
>cisco-voip mailing list
>cisco-voip@puck.nether.net
>https://puck.nether.net/mailman/listinfo/cisco-voip
>
>
| |
| Candace Holman 2005-11-16, 5:45 pm |
| Justin and Wes,
Thanks. It's true that we are using CMI for SMDI integration, so we
have some parameters
to play with. Your problem will probably resurface when you begin to
get international calls that want to go to voicemail. I think with a
VG248 you can use dial peers to translate numbers in the SMDI messages.
Unfortunately, I cannot apply a single mask to the calling party number,
it has to be different based on the source of the call. If the call is
generic PSTN, blank out the CPN, but if it is from an enterprise
exchange, keep the CPN. The legacy VM system, will look up a mailbox
with the last 5 digits of any CPN, if there is a number there. So we
have to blank out CPN for non-enterprise exchanges or there are
mismatches galore. (some algorithm, eh?)
I'm pretty sure I need to do this with variable translation patterns.
These patterns have to be applied on a call that call forwards to voice
mail through the device calling search space. A direct dial to the VM
access number doesn't require calling party number translation. So what
I'd really like is a flowchart that describes how a call routes when
there is both a translation pattern and a route pattern.
thanks,
Candace
[vbcol=seagreen]
>Candace,
>
>This sounds familiar to a problem I had with SMDI voicemail
>integration using the VG248. Anytime a call came in via PRI PSTN with
>a calling party number greater than 10 digits, the legacy voicemail
>system (an Octel) would not answer the call. It would RNA through all
>the voicemail ports. Looking further at the SMDI traces on the VG248
>I would see the VG248 passing the SMDI message, with 10+ digit calling
>party. For whatever reason, the voicemail system ignored this SMDI
>message and thus never answered the calls.
>
>I opened a case with Avaya only to be told that SMDI standards were
>either 7 or 10 digit calling party information. I was able to contact
>the telephony person at the business who was calling us with 10+ digit
>ANI. I explained the problem and they resolved the issue, so for now
>I don't have this problem - although, I'm sure it will resurface at
>some point.
>
>Candace, one thing I noticed when I researched the issue is that if
>you use CCM's CMI to handle the SMDI, via a serial cable into the
>CallManager, you have multiple parameters you can change including a
>mask for Calling Party number (There is probably something on CCO but
>I read it in the CIPT book). However, if you are using VG248 to
>handle SMDI, there isn't as many parameters and AFAIK no way to change
>CALLING party number only on the leg to voicemail.
>
>Hopefully this helps.
>
>Justin
>
>
>
| |
| Wes Sisk 2005-11-16, 5:45 pm |
| the vg248 does not use normal IOS and does not have dial-peers, so not
possible there either.
this would represent an interesting VM system.... SMDI provides specific
fields for call reason, calling party, voicemailbox, etc. treatment is
usually based first on call type and second on voicemailbox.
I have worked an issue with vg248 where international numbers were
inserted into calling party number (calltype and voicemailbox were
correctly populated) and this caused the VM system to fail because the
international numbers was too many digits long and not being truncated
by vg248. This behavior is configurable on the vg248. I believe it is
configurable in one of the CMI service parameters as well.
blanking out all callingpartynumbers before routing to VM will cause
problems - to check voicemail is a direct call to voicemail and the
calling party number (or voicemial profile mask) is used for the
voicemailbox.
If you are concerned about inbound calls routed directly to VM (users
attempting to check VM from outside the company) a translation pattern
may serve to remove calling party number before routing to VM, but i'm
not certain. I would recommend a dummy phone with voicemailprofile that
sets a specific VMBox and route that the voicemail system - then the VM
system could treat the call with the implementation specific version of
a callhandler in Unity.
/Wes
Candace Holman wrote:
[vbcol=seagreen]
> Justin and Wes,
>
> Thanks. It's true that we are using CMI for SMDI integration, so we
> have some parameters
> to play with. Your problem will probably resurface when you begin to
> get international calls that want to go to voicemail. I think with a
> VG248 you can use dial peers to translate numbers in the SMDI messages.
>
> Unfortunately, I cannot apply a single mask to the calling party
> number, it has to be different based on the source of the call. If
> the call is generic PSTN, blank out the CPN, but if it is from an
> enterprise exchange, keep the CPN. The legacy VM system, will look up
> a mailbox with the last 5 digits of any CPN, if there is a number
> there. So we have to blank out CPN for non-enterprise exchanges or
> there are mismatches galore. (some algorithm, eh?)
>
> I'm pretty sure I need to do this with variable translation patterns.
> These patterns have to be applied on a call that call forwards to
> voice mail through the device calling search space. A direct dial to
> the VM access number doesn't require calling party number
> translation. So what I'd really like is a flowchart that describes
> how a call routes when there is both a translation pattern and a route
> pattern.
>
> thanks,
> Candace
>
>
|
|
|
|
|