Re: What does 'exchange' mean and which H.245 message gets to be sent first?
Paul, H.323v4 is indeed where I got the text from, but that still leaves me clueless about what these procedures could be. You say that the sending of the TCS can force the remote EP into sending a TCS itself. I didn't think it is also the case that upon reception of a TCS, the receiving EP also has to send a TCS. It is only in this first occasion (or after an "empty TCS"), that this situation arises. What is your interpretation of "capability exchange" (I.e. the sending of the capability set only or the mutual exchange of capabilities)? Regards, Frank plong@IPDIALOG.COM@SMTP@MAILBAG.INTEL.COM on 03/04/2001 19:18:54 Please respond to plong@IPDIALOG.COM@SMTP Sent by: ITU-SG16@MAILBAG.INTEL.COM To: ITU-SG16@MAILBAG.INTEL.COM@SMTP cc: Subject: Re: What does 'exchange' mean and which H.245 message gets to be sent first? Classification: Frank, 1. I have no idea what those failing procedures would be. Maybe it's a v4 thing. 2. Yes, that is correct. Because the TCS must be the first message, if an EP receives a TCS, it must sends its own TCS (if it has not already) before it responds to the TCS with, typically, TCSAck. This is how an EP can force the remote EP into performing cap exchange. Paul Long ipDialog, Inc. -----Original Message----- From: Mailing list for parties associated with ITU-T Study Group 16 [mailto:ITU-SG16@mailbag.cps.INTEL.COM]On Behalf Of Frank Derks Sent: Tuesday, April 03, 2001 10:33 AM To: ITU-SG16@mailbag.cps.INTEL.COM Subject: What does 'exchange' mean and which H.245 message gets to be sent first? H.323 talks about capability exchange, but it is not clear to me whether exchange really means that communicating endpoints send their capability sets to _each other_ or that one endpoint sends its capability set. An example of this can be found in the below text. It talks about succesfully completing capability exchange, but does this means that the initiating endpoint has sent its capability set and that this was acknowledge by the receiver or does it mean that it has also received the capability set of the other party? 8.2/H.323 states: "Endpoint system capabilities are exchanged by transmission of the H.245 terminalCapabilitySet message. This capability message shall be the first H.245 message sent unless the endpoint is indicating that it understands the parallelH245Control field (see section 8.2.4). If prior to successful completion of terminal capability exchange, any other procedure fails, (i.e. rejected, not understood, not supported) then the initiating endpoint should initiate and successfully complete terminal capability exchange before attempting any other procedure. " The second sentence mandates that the terminalCapabilitySet message be the first message sent (given that "parallelH245Control" is not used). Which "failing procedures" are then being referred to in the third sentence? If terminalCapabilitySet has to be the first message to be sent, does this mean that an endpoint that has not sent a terminalCapabilitySet message yet and receives this message, it first has to send a terminalCapabilitySet message before acknowledging the received capability set? Frank ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ For help on this mail list, send "HELP ITU-SG16" in a message to listserv@mailbag.intel.com ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ For help on this mail list, send "HELP ITU-SG16" in a message to listserv@mailbag.intel.com
Frank, It's not that an EP must send TCS _every time_ it receives a TCS, only the first time if it has not already sent it for this session. It just needs to satisfy the TCS-first rule (before any other message, including acks). That's all. "Capability exchange" may be overloaded. Strictly speaking, I think it refers to the true exchange at the beginning of the H.245 session where both EPs send their caps to the other EP. This may be referred to more specifically as "THE Capability Exchange." When an EP unilaterally sends a subsequent TCS, it could also be said to be performing a capability exchange, but this is misleading. It is not "exchanging" capabilities, per se, because the remote EP is not required to also send its caps. Paul Long ipDialog, Inc. -----Original Message----- From: Mailing list for parties associated with ITU-T Study Group 16 [mailto:ITU-SG16@mailbag.cps.INTEL.COM]On Behalf Of Frank Derks Sent: Tuesday, April 03, 2001 1:15 PM To: ITU-SG16@mailbag.cps.INTEL.COM Subject: Re: What does 'exchange' mean and which H.245 message gets to be sent first? Paul, H.323v4 is indeed where I got the text from, but that still leaves me clueless about what these procedures could be. You say that the sending of the TCS can force the remote EP into sending a TCS itself. I didn't think it is also the case that upon reception of a TCS, the receiving EP also has to send a TCS. It is only in this first occasion (or after an "empty TCS"), that this situation arises. What is your interpretation of "capability exchange" (I.e. the sending of the capability set only or the mutual exchange of capabilities)? Regards, Frank plong@IPDIALOG.COM@SMTP@MAILBAG.INTEL.COM on 03/04/2001 19:18:54 Please respond to plong@IPDIALOG.COM@SMTP Sent by: ITU-SG16@MAILBAG.INTEL.COM To: ITU-SG16@MAILBAG.INTEL.COM@SMTP cc: Subject: Re: What does 'exchange' mean and which H.245 message gets to be sent first? Classification: Frank, 1. I have no idea what those failing procedures would be. Maybe it's a v4 thing. 2. Yes, that is correct. Because the TCS must be the first message, if an EP receives a TCS, it must sends its own TCS (if it has not already) before it responds to the TCS with, typically, TCSAck. This is how an EP can force the remote EP into performing cap exchange. Paul Long ipDialog, Inc. -----Original Message----- From: Mailing list for parties associated with ITU-T Study Group 16 [mailto:ITU-SG16@mailbag.cps.INTEL.COM]On Behalf Of Frank Derks Sent: Tuesday, April 03, 2001 10:33 AM To: ITU-SG16@mailbag.cps.INTEL.COM Subject: What does 'exchange' mean and which H.245 message gets to be sent first? H.323 talks about capability exchange, but it is not clear to me whether exchange really means that communicating endpoints send their capability sets to _each other_ or that one endpoint sends its capability set. An example of this can be found in the below text. It talks about succesfully completing capability exchange, but does this means that the initiating endpoint has sent its capability set and that this was acknowledge by the receiver or does it mean that it has also received the capability set of the other party? 8.2/H.323 states: "Endpoint system capabilities are exchanged by transmission of the H.245 terminalCapabilitySet message. This capability message shall be the first H.245 message sent unless the endpoint is indicating that it understands the parallelH245Control field (see section 8.2.4). If prior to successful completion of terminal capability exchange, any other procedure fails, (i.e. rejected, not understood, not supported) then the initiating endpoint should initiate and successfully complete terminal capability exchange before attempting any other procedure. " The second sentence mandates that the terminalCapabilitySet message be the first message sent (given that "parallelH245Control" is not used). Which "failing procedures" are then being referred to in the third sentence? If terminalCapabilitySet has to be the first message to be sent, does this mean that an endpoint that has not sent a terminalCapabilitySet message yet and receives this message, it first has to send a terminalCapabilitySet message before acknowledging the received capability set? Frank ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ For help on this mail list, send "HELP ITU-SG16" in a message to listserv@mailbag.intel.com ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ For help on this mail list, send "HELP ITU-SG16" in a message to listserv@mailbag.intel.com ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ For help on this mail list, send "HELP ITU-SG16" in a message to listserv@mailbag.intel.com
participants (2)
-
Frank Derks -
Paul Long