[Home]

Summary:ASTERISK-03792: [patch] implement sending of 'early'-state in NOTIFY-Messages
Reporter:maik (maik)Labels:
Date Opened:2005-03-28 06:43:38.000-0600Date Closed:2005-04-01 00:25:24.000-0600
Priority:MajorRegression?No
Status:Closed/CompleteComponents:Channels/chan_sip/Subscriptions
Versions:Frequency of
Occurrence
Related
Issues:
Environment:Attachments:( 0) sip-notify-early.20050331.diff.txt
( 1) sip-notify-early.diff.txt
( 2) sip-notify-early-with-option.20050331.diff.txt
( 3) sip-notify-early-with-option.diff.txt
Description:This patch implements the sending of NOTIFY-messages with state 'early' when the subscribed extension is ringing. It was tested with a snom 360 but it should also work at least with other snom models (the LED should blink when the extension is ringing).

****** ADDITIONAL INFORMATION ******

Disclaimer on file
Comments:By: Kevin P. Fleming (kpfleming) 2005-03-28 11:49:47.000-0600

Shouldn't RINGING_AND_INUSE set the light on, rather than blinking?

By: maik (maik) 2005-03-28 12:04:23.000-0600

I discussed this with several people and most of them said that blinking would be the better option. It depends on the situation I think so maybe it should be configurable...

By: Kevin P. Fleming (kpfleming) 2005-03-28 12:10:06.000-0600

I guess ringing makes sense, now that I think about it again. If none of the devices for that extension are configured to support call waiting, they'll just bounce the call back, so this doesn't matter. If they are, then the light will go from solid->blinking->solid, which is logical.

By: maik (maik) 2005-03-30 18:06:54.000-0600

I did not have the time to implement the config-option yet but I just uploaded a new version of the patch for latest CVS Head.

By: maik (maik) 2005-03-30 19:15:32.000-0600

I could not sleep so I had some time to implement the config-option.
Unfortunately I don't have a phone here to test if it really works but I think I will have time to do that tomorrow.

Maybe someone has an idea for a better name because I don't think 'notifyringingandinuseasinuse' is what we want to have in a config file. :)
I'm too tired now to come up with something better.

By: Kevin P. Fleming (kpfleming) 2005-03-30 21:54:18.000-0600

You are correct, that config option name is way too long.

Is there any way to determine from the device whether it 'supports' early or not? If so, this could be automatically enabled when the device claims to support it.

If that can't be done, this needs some sort of name that is easier to use, but I can't come up with a good shorthand name for it. It also needs to be specifiable at the peer level, not just globally.

By: maik (maik) 2005-03-30 21:59:56.000-0600

I think every device that supports the other states also supports early. All the states are defined in the same document (draft-ietf-sipping-dialog-package-05.txt).

By: Kevin P. Fleming (kpfleming) 2005-03-30 22:06:23.000-0600

In that case, your patch is just more fully implementing the existing 'standard', so it probably doesn't need a config option at all.

By: maik (maik) 2005-03-30 22:26:35.000-0600

The config option was just for people who think like you did when you posted the first time. So I think a global option could make sense.

By: Kevin P. Fleming (kpfleming) 2005-03-30 22:33:06.000-0600

Well, I'll let somebody else make the final decision there, but you'll have to come up with a better name for the option... something like "notifyringing" might be OK.

By: maik (maik) 2005-03-30 22:35:44.000-0600

OK. I'll change the option name and try to test it today.

By: Olle Johansson (oej) 2005-03-31 01:32:51.000-0600

We need to merge this with ASTERISK-3564. I'll take a look at it during the weekend and see how they work together.

By: maik (maik) 2005-03-31 11:32:09.000-0600

I just uploaded a new version (sip-notify-early-with-option.20050331.diff.txt) which fixes some bugs and has been tested with a snom 360.

oej: I'll have a look at your patch now.

By: maik (maik) 2005-03-31 12:43:28.000-0600

I just merged my own changes into oej's patch (3644) so I guess this bug can be closed.

By: Olle Johansson (oej) 2005-04-01 00:25:24.000-0600

--merged into another patch, not for 1.0