Menu
▾
▴
turnserver-devel — Discussions on development of TurnServer
You can subscribe to this list here.
| 2008 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
(1) |
Aug
(2) |
Sep
(3) |
Oct
(1) |
Nov
(3) |
Dec
(3) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2009 |
Jan
|
Feb
(1) |
Mar
(1) |
Apr
(2) |
May
|
Jun
|
Jul
|
Aug
(3) |
Sep
|
Oct
|
Nov
(15) |
Dec
(3) |
| 2010 |
Jan
(2) |
Feb
(1) |
Mar
|
Apr
(5) |
May
(1) |
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(1) |
Dec
(2) |
| 2011 |
Jan
(4) |
Feb
(1) |
Mar
(2) |
Apr
|
May
(3) |
Jun
(2) |
Jul
(2) |
Aug
|
Sep
|
Oct
(10) |
Nov
(14) |
Dec
|
| 2012 |
Jan
|
Feb
(4) |
Mar
(5) |
Apr
(4) |
May
(4) |
Jun
(2) |
Jul
(11) |
Aug
|
Sep
|
Oct
(6) |
Nov
(14) |
Dec
(4) |
| 2013 |
Jan
(6) |
Feb
|
Mar
(2) |
Apr
|
May
(3) |
Jun
(2) |
Jul
|
Aug
|
Sep
(1) |
Oct
|
Nov
|
Dec
(3) |
| 2014 |
Jan
(4) |
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| 2015 |
Jan
(1) |
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| 2016 |
Jan
(1) |
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: Sébastien V. <vi...@cl...> - 2008-12-17 13:49:45
|
Hello all, Thanks to Paul Mackell, he fix the CRC32 generation which now is correct. So I made this "bug fix" release today, I have also integrated a drop privileges system if TurnServer is launched as root / setuid root. Regards, -- Sebastien Vincent |
|
From: Sébastien V. <vi...@cl...> - 2008-12-15 10:41:15
|
Dear all, Our apologies if you receive multiple copies of this announcement. We are glad to announce the release of TurnServer version 0.2, the OpenSource TURN server. This new version supports RFC5389 (STUN), latest turn-12 and turn-ipv6-05 drafts. We also refactor some code parts and add other things like setting DF flag on Linux and quotas / IP restriction stuff. Here is the ChangeLog: - Add attributes from turn-12 draft; - Add authentication on all TURN responses when possible; - Pre-calculate account MD5 hash; - Re-organize some code (message-integrity calculation, ...); - Alternate behavior for DF flag; - Separate TCP and TLS sockets; - Correctly relayed data from peer to client in TLS; - Multiple XOR-PEER-ADDRESS support in CreatePermission; - Update wireshark parser (packet-stun2.c); - Additionnal configuration parameters; - Allocation number and bandwidth quota; - Address / port restrictions; - Reconstruct fragmented TCP messages support; - Prevent allocation hijacking support; - Add logs support via syslog. TurnServer code can be retrieved from the download page: http://sourceforge.net/project/showfiles.php?group_id=234986 The mailing-lists are available at: http://lists.sourceforge.net/mailman/listinfo/turnserver-devel http://lists.sourceforge.net/mailman/listinfo/turnserver-users As usual, any feedback is welcome. Best regards, -- Sebastien Vincent http://www.turnserver.org/ http://sourceforge.net/projects/turnserver/ |
|
From: Sébastien V. <vi...@cl...> - 2008-12-02 15:38:26
|
Hello all, Now SVN trunk repository holds the 0.2 version of TurnServer (formerly known as turnserver-draft-11). It will be soon ready for a release and will support turn-12 (yes we are already at revision 12) specification. The last version that support turn-09 is turnserver-0.1.3.1 which is in SVN tags repository. Regards, -- Sebastien Vincent |
|
From: Sébastien V. <vi...@cl...> - 2008-11-29 08:03:19
|
Hello Alexander, Sorry for this _very_ late reply. I have implemented the messages reconstruction on TCP stream in TurnServer. So if you could test it in real condition and report me any problems. Please note that it is in 0.2 branch (svn co https://turnserver.svn.sourceforge.net/svnroot/turnserver/branches/turnserver-draft-11) and that this version support only turn-11 draft. Regards, -- Seb Alexander Pevzner a écrit : > Hi, > > I'm trying to use the turnserver as a solution for NAT traversal for a > very restricted network environment with only TCP traffic allowed by > firewall. > > If TCP is being used as a transport between client and server, the > following problem occurs. If client sends more that one TURN message to > the server, only the first message is handled by server normally, all > other messages cause server to print "Bad STUN / TURN message" diagnostics. > > Looking to a server code I came to conclusion that this is due to a way > how TCP reception is implemented in the server. > > Server calls recv() for a TCP socket and assumes that the returned data > is a TURN message. > > This assumption is not correct. TCP is a stream rather that message > protocol, it doesn't preserve message boundaries. It means that recv() > may return more that one message at once, or message may be split > between consecutive recv()s and so on. > > The only correct way to deal with TCP is to reconstruct message > boundaries based on a message size transmitted in a message headers. > > Data from TCP socket should go to some temporary buffer, and only when > entire message is available it needs to be processed. > > -- > Hope it helps, > Wishes, Alexander Pevzner (pz...@pz...) > > ------------------------------------------------------------------------- > This SF.Net email is sponsored by the Moblin Your Move Developer's challenge > Build the coolest Linux based applications with Moblin SDK & win great prizes > Grand prize is a trip for two to an Open Source event anywhere in the world > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > _______________________________________________ > Turnserver-devel mailing list > Tur...@li... > https://lists.sourceforge.net/lists/listinfo/turnserver-devel > > -- Sébastien Vincent Network Engineer at University Louis Pasteur, Strasbourg Phone : +33 390244586 Address : LSIIT - Boulevard Sébastien Brant BP 10413 67412 Illkirch CEDEX France |
|
From: Sébastien V. <vi...@cl...> - 2008-11-20 18:05:53
|
Hello all, TurnServer version 0.1.3 will be released november 22th. This will be the last version which supports turn-09 draft, so if you develop a turn-09 client I recommend you to switch on new turn-11 draft specification :). For those who cannot wait to test it, check our SVN repository: svn co https://turnserver.svn.sourceforge.net/svnroot/turnserver/branches/trunk/ Next month, I will release the 0.2 version which support the newer turn-11 draft and have some code refactoring. If you have the time, please test the 0.2 branches available on our SVN repository: svn co https://turnserver.svn.sourceforge.net/svnroot/turnserver/branches/turnserver-draft-11 Regards, -- Sebastien Vincent |
|
From: Sébastien V. <vi...@cl...> - 2008-11-03 20:18:46
|
Hello all, I begin to upgrade TurnServer to the latest TURN draft : http://tools.ietf.org/html/draft-ietf-behave-turn-11. I have commited on turnserver-draft-11 branche: svn co https://turnserver.svn.sourceforge.net/svnroot/turnserver/branches/turnserver-draft-11 turnserver-draft-11 Things already done: - Add attributes from turn-11 draft; - Add authentication on all TURN responses when possible; - Pre-calculate account MD5 hash; - Re-organize some code (message-integrity calculation, ...); - Alternate behavior for DF flag; - Separate TCP and TLS sockets; - Correctly relayed data from peer to client in TLS; - Typos. Things that need to be done: - Support multiple XOR-PEER-ADDRESS - Quotas on allocation per user; - IP range restriction (CreatePermission); - List OS that support or not setting DF flag. Note that this version is experimental. So I encourage you to test this version and give feedbacks (something not working / contrary to the draft, missing functionnality from the draft, ...). Regards, -- Sebastien Vincent |
|
From: Alexander P. <pz...@pz...> - 2008-10-14 16:20:53
|
Hi, I'm trying to use the turnserver as a solution for NAT traversal for a very restricted network environment with only TCP traffic allowed by firewall. If TCP is being used as a transport between client and server, the following problem occurs. If client sends more that one TURN message to the server, only the first message is handled by server normally, all other messages cause server to print "Bad STUN / TURN message" diagnostics. Looking to a server code I came to conclusion that this is due to a way how TCP reception is implemented in the server. Server calls recv() for a TCP socket and assumes that the returned data is a TURN message. This assumption is not correct. TCP is a stream rather that message protocol, it doesn't preserve message boundaries. It means that recv() may return more that one message at once, or message may be split between consecutive recv()s and so on. The only correct way to deal with TCP is to reconstruct message boundaries based on a message size transmitted in a message headers. Data from TCP socket should go to some temporary buffer, and only when entire message is available it needs to be processed. -- Hope it helps, Wishes, Alexander Pevzner (pz...@pz...) |
|
From: Sébastien V. <vi...@cl...> - 2008-09-04 06:48:53
|
Hum, in facts today (September 4th), sourceforge.net servers are in maintenance state for one day (SVN write operation will not work) so the patch will be in SVN tomorrow. -- Sebastien Vincent Sébastien Vincent a écrit : > Hello Guy, > > Thanks for reporting this bug. > > Patch (a little bit modified) has been commited to the SVN. > > Regards, > -- > Sebastien Vincent > > Guy Corem a écrit : > >> Hi Sebastien, >> >> I've tested with "daemon = true", and on my ubuntu 8.04 system, the server daemon seems stuck. >> I've found out that by not closing stdin, the server is working fine. >> I've attached suggested patch... >> >> >> ------------------------------------------------------------------------ >> >> ------------------------------------------------------------------------- >> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge >> Build the coolest Linux based applications with Moblin SDK & win great prizes >> Grand prize is a trip for two to an Open Source event anywhere in the world >> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> Turnserver-devel mailing list >> Tur...@li... >> https://lists.sourceforge.net/lists/listinfo/turnserver-devel >> > > ------------------------------------------------------------------------- > This SF.Net email is sponsored by the Moblin Your Move Developer's challenge > Build the coolest Linux based applications with Moblin SDK & win great prizes > Grand prize is a trip for two to an Open Source event anywhere in the world > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > _______________________________________________ > Turnserver-devel mailing list > Tur...@li... > https://lists.sourceforge.net/lists/listinfo/turnserver-devel > > -- Sébastien Vincent Network Engineer at University Louis Pasteur, Strasbourg Phone : +33 390244586 Address : LSIIT - Boulevard Sébastien Brant BP 10413 67412 Illkirch CEDEX France |
|
From: Sébastien V. <vi...@cl...> - 2008-09-04 06:45:44
|
Hello Guy, Thanks for reporting this bug. Patch (a little bit modified) has been commited to the SVN. Regards, -- Sebastien Vincent Guy Corem a écrit : > Hi Sebastien, > > I've tested with "daemon = true", and on my ubuntu 8.04 system, the server daemon seems stuck. > I've found out that by not closing stdin, the server is working fine. > I've attached suggested patch... > > > ------------------------------------------------------------------------ > > ------------------------------------------------------------------------- > This SF.Net email is sponsored by the Moblin Your Move Developer's challenge > Build the coolest Linux based applications with Moblin SDK & win great prizes > Grand prize is a trip for two to an Open Source event anywhere in the world > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > ------------------------------------------------------------------------ > > _______________________________________________ > Turnserver-devel mailing list > Tur...@li... > https://lists.sourceforge.net/lists/listinfo/turnserver-devel |
|
From: Guy C. <gu...@no...> - 2008-09-03 21:51:50
|
Hi Sebastien, I've tested with "daemon = true", and on my ubuntu 8.04 system, the server daemon seems stuck. I've found out that by not closing stdin, the server is working fine. I've attached suggested patch... -- Guy Corem Nop90 (2007) Ltd. Mobile: +972-50-7321946 Fax: 153-50-7321946 email: gu...@no... web: http://www.nop90.com |
|
From: Sébastien V. <vi...@cl...> - 2008-08-24 19:51:36
|
Hello Guy, Thanks for reporting ! Patch applied. Regards -- Sebastien Guy Corem a écrit : > Hi Sebastien, > > When testing interoprability with numb (http://numb.viagenie.ca/), > I've found out that test_client_udp got stuck in infinite loop. > numb is returning the obsolete BANDWIDTH. > > I believe that I've spotted the bug. > See attached suggest patch. > > Thanks, > Guy > > ------------------------------------------------------------------------ > > ------------------------------------------------------------------------- > This SF.Net email is sponsored by the Moblin Your Move Developer's challenge > Build the coolest Linux based applications with Moblin SDK & win great prizes > Grand prize is a trip for two to an Open Source event anywhere in the world > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > ------------------------------------------------------------------------ > > _______________________________________________ > Turnserver-devel mailing list > Tur...@li... > https://lists.sourceforge.net/lists/listinfo/turnserver-devel > -- Sébastien Vincent Network Engineer at University Louis Pasteur, Strasbourg Phone : +33 390244586 Address : LSIIT - Boulevard Sébastien Brant BP 10413 67412 Illkirch CEDEX France |
|
From: Guy C. <vc...@gm...> - 2008-08-24 17:38:38
|
Hi Sebastien, When testing interoprability with numb (http://numb.viagenie.ca/), I've found out that test_client_udp got stuck in infinite loop. numb is returning the obsolete BANDWIDTH. I believe that I've spotted the bug. See attached suggest patch. Thanks, Guy |
|
From: Sebastien V. <seb...@cp...> - 2008-07-26 15:57:46
|
Welcome to the "devel" mailing list. |
2 messages has been excluded from this view by a project administrator.
