Commit Graph

24 Commits

Author SHA1 Message Date
kurtis.heimerl
d2871d763f r4161 in private:
SIPEngine::instigator now based on a flag instead of a string comparison.

For some reason that we need to understand, string comparisons on oSIP messages are really not reliable.

git-svn-id: http://wush.net/svn/range/software/public/openbts/trunk@4173 19bc5d8c-e614-43d4-8b26-e1612bc8e597
2012-09-24 20:02:32 +00:00
kurtis.heimerl
059eea60af Set state to fail instead of throwing a timeout in MOSMSWaitForSubmit
git-svn-id: http://wush.net/svn/range/software/public/openbts/trunk@3925 19bc5d8c-e614-43d4-8b26-e1612bc8e597
2012-07-24 05:55:51 +00:00
kurtis.heimerl
0623af8295 Fixed null pointer in MOSMSSubmit.
git-svn-id: http://wush.net/svn/range/software/public/openbts/trunk@3923 19bc5d8c-e614-43d4-8b26-e1612bc8e597
2012-07-24 05:29:37 +00:00
kurtis.heimerl
f754059314 We no longer repeat INVITEs when we get a 100 TRYING.
git-svn-id: http://wush.net/svn/range/software/public/openbts/trunk@3883 19bc5d8c-e614-43d4-8b26-e1612bc8e597
2012-07-11 00:06:25 +00:00
kurtis.heimerl
0de2258acf removed notice about not caring about 100 trying messages. RFCs be damned, the proxies use them.
git-svn-id: http://wush.net/svn/range/software/public/openbts/trunk@3857 19bc5d8c-e614-43d4-8b26-e1612bc8e597
2012-06-28 06:10:45 +00:00
kurtis.heimerl
f3e57f7b19 SIP MESSAGES now wait when they receive a 100 TRYING, rather than repeating themselves.
git-svn-id: http://wush.net/svn/range/software/public/openbts/trunk@3856 19bc5d8c-e614-43d4-8b26-e1612bc8e597
2012-06-28 06:10:41 +00:00
kurtis.heimerl
562dd0d7fb Fixes #991 in public. We now appropriately handle rejections of our SIP MESSAGE content
git-svn-id: http://wush.net/svn/range/software/public/openbts/trunk@3812 19bc5d8c-e614-43d4-8b26-e1612bc8e597
2012-06-19 21:33:26 +00:00
kurtis.heimerl
bbc25c6a10 Fixed #932 and #905 in public. OpenBTS now responds negatively to SIP
re-invites, though the specific effect of this depends on the client
 sending the re-invite. In the case of FS, it causes the RTP stream to
 break. We also now complain when invites time out, sending an alert to the
 user about the proxy status.

git-svn-id: http://wush.net/svn/range/software/public/openbts/trunk@3810 19bc5d8c-e614-43d4-8b26-e1612bc8e597
2012-06-19 20:22:20 +00:00
kurtis.heimerl
c084181a75 attempting to fix #937 in public, basically forgot to check if sip_method exists.
git-svn-id: http://wush.net/svn/range/software/public/openbts/trunk@3700 19bc5d8c-e614-43d4-8b26-e1612bc8e597
2012-05-31 02:44:12 +00:00
kurtis.heimerl
55d1bf25a4 $ dropped by the network by retransmitting.
git-svn-id: http://wush.net/svn/range/software/public/openbts/trunk@3677 19bc5d8c-e614-43d4-8b26-e1612bc8e597
2012-05-28 01:07:16 +00:00
kurtis.heimerl
3d4ec90bd1 Fixes #734 in public, , as well as rearchitects a number of SIP
control flows. Basically, there was a small bug where I forgot to listen
 for an OK, which caused transactions to stick around too long and fuck
 everything up. That was quickly fixed, but my compulsion required me to
 reorganize the code to avoid such errors in the future. I implemented
 another new feature as well, we now respond with a 480 Temporarily
 Unavailable message (rather than cancel) when canceling an INCOMING call.
 The prior behavior was incorrect.

git-svn-id: http://wush.net/svn/range/software/public/openbts/trunk@3264 19bc5d8c-e614-43d4-8b26-e1612bc8e597
2012-03-05 03:53:57 +00:00
kurtis.heimerl
a954b789f0 Fixes #499 in public. If our ACK was missed, the switch would resend an OK and we didn't respond to it. We do now.
git-svn-id: http://wush.net/svn/range/software/public/openbts/trunk@3237 19bc5d8c-e614-43d4-8b26-e1612bc8e597
2012-02-25 21:46:49 +00:00
kurtis.heimerl
b0f5d61bf5 Fixes #711 in private. We now check RTP ports before allocation, to ensure they're not already in use.
git-svn-id: http://wush.net/svn/range/software/public/openbts/trunk@3225 19bc5d8c-e614-43d4-8b26-e1612bc8e597
2012-02-24 07:59:11 +00:00
kurtis.heimerl
36b618f773 Fixes r168 in public. This involves the handling of "481 Call leg/transaction does not exist. Basically, we occasionally get this in response to a BYE or CANCEL (as it's been closed on the switch side). This was almost certainly fixed in a previous commit. I've updated the log message to clarify that the connection is cleared when we receive a 481.
git-svn-id: http://wush.net/svn/range/software/public/openbts/trunk@3150 19bc5d8c-e614-43d4-8b26-e1612bc8e597
2012-02-05 02:53:38 +00:00
kurtis.heimerl
041e211b2c Fixes #501 in the public release.
git-svn-id: http://wush.net/svn/range/software/public/openbts/trunk@3037 19bc5d8c-e614-43d4-8b26-e1612bc8e597
2012-01-12 08:13:03 +00:00
kurtis.heimerl
64656ed5b5 Fixes #34 in public release. We now have different states for the MTDCanceling and canceled calls now end in a canceled state. Also a few other hardening bug fixes.
git-svn-id: http://wush.net/svn/range/software/public/openbts/trunk@3018 19bc5d8c-e614-43d4-8b26-e1612bc8e597
2012-01-10 22:57:53 +00:00
kurtis.heimerl
c4c41c65b6 Fixed Bug #172, OpenBTS now sends CANCEL messages when there is no active call (instead of BYE)
git-svn-id: http://wush.net/svn/range/software/public/openbts/trunk@3006 19bc5d8c-e614-43d4-8b26-e1612bc8e597
2012-01-09 01:01:12 +00:00
kurtis.heimerl
c614ad4a4d Small typo fix
git-svn-id: http://wush.net/svn/range/software/public/openbts/trunk@2996 19bc5d8c-e614-43d4-8b26-e1612bc8e597
2012-01-07 09:41:27 +00:00
kurtis.heimerl
fb00b2c122 Bug #173-public. Now appropriately responds to a SIP CANCEL with a SIP OK. Still won't respond if that OK is dropped.
git-svn-id: http://wush.net/svn/range/software/public/openbts/trunk@2995 19bc5d8c-e614-43d4-8b26-e1612bc8e597
2012-01-07 09:41:24 +00:00
kurtis.heimerl
0118e2495f bug #238-private. OpenBTS segfaults when receving an empty SMS destined to go out as text/plain. It no longer segfaults, but does not successfully send the message.
git-svn-id: http://wush.net/svn/range/software/public/openbts/trunk@2969 19bc5d8c-e614-43d4-8b26-e1612bc8e597
2012-01-05 02:25:08 +00:00
kurtis.heimerl
410a0cc329 Alexander's change. Fixes MTC bug by supporting 64bit processors in Timeval. Math was bad.
git-svn-id: http://wush.net/svn/range/software/public/openbts/trunk@2835 19bc5d8c-e614-43d4-8b26-e1612bc8e597
2011-12-17 00:03:24 +00:00
kurtis.heimerl
3129fc2bca more documentation
git-svn-id: http://wush.net/svn/range/software/public/openbts/trunk@2823 19bc5d8c-e614-43d4-8b26-e1612bc8e597
2011-12-15 06:30:52 +00:00
kurtis.heimerl
26dfc03365 Fixing Thomas's MTC receive bug. We don't have a ticket for it, but there seemed to be a timing inversion deadlock in Interthread.h. We mostly duck it, instead of solving it.
git-svn-id: http://wush.net/svn/range/software/public/openbts/trunk@2822 19bc5d8c-e614-43d4-8b26-e1612bc8e597
2011-12-15 06:23:39 +00:00
dburgess
508962442c Putting the actual OpenBTS P2.8 source code into the public SVN branch.
git-svn-id: http://wush.net/svn/range/software/public/openbts/trunk@2242 19bc5d8c-e614-43d4-8b26-e1612bc8e597
2011-10-07 02:40:51 +00:00