Travis Cross
|
9b611f0175
|
Spelling fix
|
2013-06-07 07:26:29 +00:00 |
|
William King
|
66652ec198
|
how...?
|
2013-06-07 07:26:28 +00:00 |
|
William King
|
b214f20768
|
Since there isn't neutral ground, on the rtmp state callback we actually have to force the write lock of the session, but we only need a try lock on removing the session from the rsession->session_hash. This removes a deadlock caused when we had to enforce with locks that the rsession couldn't be destroyed if the FS session was in the middle of a hangup.
|
2013-06-07 07:26:28 +00:00 |
|
William King
|
ce9bf959e9
|
Seems the session wasn't coming to the hangup in a locked state.
|
2013-06-07 07:26:27 +00:00 |
|
William King
|
1ff548f24c
|
Seems having a read lock still prevents a write lock to be taken, even if the thread that is trying to get the write lock currently has the read lock.
|
2013-06-07 07:26:27 +00:00 |
|
William King
|
8d370ff6e6
|
Include the write lock since we're changing the tech_pvt
|
2013-06-07 07:26:26 +00:00 |
|
Jeff Lenk
|
ec44f6aec0
|
FS-5167
|
2013-06-07 07:26:11 +00:00 |
|
William King
|
b424ad303d
|
FS-5440: don't assert on bad input rtmp data
|
2013-06-07 05:42:44 +00:00 |
|
Giovanni Maruzzelli
|
e8d3712d56
|
test 2, please ignore
|
2013-06-07 05:19:02 +00:00 |
|
root
|
a2c44f171b
|
test, please ignore
|
2013-06-07 05:18:58 +00:00 |
|
Anthony Minessale
|
102377ebf9
|
FS-5493 --resolve
|
2013-06-06 12:59:08 -05:00 |
|
Anthony Minessale
|
50b68f2f85
|
add some device-state mechinism to FS to allow tracking of device-specific states where they may have more than one call from the same device
|
2013-06-05 11:19:53 -05:00 |
|
Nathan Neulinger
|
2e66012e96
|
Change to correct prompt on outgoing call, eliminate UNDEFINED! message
|
2013-05-31 15:10:44 -05:00 |
|
Nathan Neulinger
|
d39086892f
|
Allow skinny speed dials to work when already off-hook
|
2013-05-31 15:10:44 -05:00 |
|
Nathan Neulinger
|
123d0865fb
|
FS-5431: --resolve fix socket creation loop for mod_skinny to avoid segfault
|
2013-05-31 15:10:43 -05:00 |
|
Nathan Neulinger
|
32bddaa6cd
|
FS-5429: --resolve add minimal idivert support for skinny
|
2013-05-31 15:10:43 -05:00 |
|
Nathan Neulinger
|
582e3699f6
|
FS-5379: --resolve fix skinny phone codec agreement issues due to incorrect hardwired ptime, update field names for future coding to be more descriptive
|
2013-05-31 15:10:43 -05:00 |
|
Nathan Neulinger
|
da5c2a6aef
|
FS-5225: --resolve add ability to dial by pressing line button
|
2013-05-31 15:10:43 -05:00 |
|
Nathan Neulinger
|
c33eb7412b
|
FS-5267: --resolve avoid null deref if sessions missing
|
2013-05-31 15:10:43 -05:00 |
|
Mathieu Parent
|
ab7635e9bc
|
Add IPs to mod_skinny recv/send event messages
FS-5268 --resolve
thanks to Nathan Neulinger
|
2013-05-31 15:10:43 -05:00 |
|
Mathieu Parent
|
4a981b336a
|
Don't pretend we support skinny-wait timeout (See #FS 477)
|
2013-05-31 15:10:42 -05:00 |
|
Mathieu Parent
|
b36ad82da7
|
Don't create a new session when pressing speeddial
Fixes #FS-5259 (Thanks to Nathan Neulinger for the patch)
|
2013-05-31 15:10:42 -05:00 |
|
Brian West
|
83c6f4250b
|
add kirk wireless servers to safe as they do not do rport
|
2013-05-31 11:56:53 -05:00 |
|
Travis Cross
|
acc8eb5cb6
|
Remove unused variable in mod_rtmp
Unused variables break the build with recent versions of gcc and
clang.
|
2013-05-26 05:29:21 +00:00 |
|
William King
|
dfbc175776
|
Let's bit a bit safer when dereferencing pointers.
|
2013-05-24 20:03:19 -07:00 |
|
William King
|
6c3062a768
|
Removing logically dead code
|
2013-05-24 19:59:20 -07:00 |
|
William King
|
6320042f6b
|
Another location where mod_rtmp would leak memory.
|
2013-05-24 19:45:24 -07:00 |
|
William King
|
83ae82fd2d
|
Fix improper handling of a double linked list that could have caused a memory leak.
|
2013-05-24 19:42:02 -07:00 |
|
William King
|
b3aea78730
|
Appears to have been a copy paste error because this doesn't do what the api claims, and it also leaks the reference to the new node.
|
2013-05-24 19:35:43 -07:00 |
|
William King
|
8a743cb2f3
|
Fixing a memory leak. name needs to be freed since it's being strdup'd lower.
|
2013-05-24 19:31:40 -07:00 |
|
Anthony Minessale
|
22076b7b6f
|
FS-5453 --resolve
|
2013-05-24 08:13:55 -05:00 |
|
Anthony Minessale
|
9c98829aab
|
FS-5453 --resolve this is a different issue, we just have to hangup the loopback channel so the loopback and the sip don't both execute the extension
|
2013-05-23 09:59:18 -05:00 |
|
Anthony Minessale
|
3d2eb20a28
|
FS-5453 --resolve
|
2013-05-23 08:06:52 -05:00 |
|
Anthony Minessale
|
1b6860d8e3
|
FS-5453 --resolve
|
2013-05-21 12:02:12 -05:00 |
|
William King
|
79e3de3e55
|
full_contact must be freed by sofia. There isn't a chance of this having blown anything up, but Coverity is right that this should be initialized with NULL to be safe against future refactors.
|
2013-05-18 19:13:44 -07:00 |
|
William King
|
97286e10cd
|
Possible leak if someone were to try and add extra headers to a channel that had either an inbound info on that session or the session were to be deflected. I don't see a way that these extra headers could be added outside of a C module subscribing to the session states, but just to make sure I'm adding these so that just like other extra header locations it's cleaned up.
|
2013-05-18 17:01:27 -07:00 |
|
William King
|
14f48a6a7a
|
Alice meet rabit hole. There is a leak here on every invite that comes back from a status 180 when the other side is not FS, but you have presence enabled. Thanks to Coverity scan for location this condition. This 'might' not have been leaking if sofia was automatically freeing this when the sofia object was destroyed, but all other locations where sip_header_as_string where called they were all followed by a su_free, so I assume not.
|
2013-05-18 16:36:41 -07:00 |
|
William King
|
d61d395e8d
|
If the mod_sofia global presense hash is null, then make sure to free the matches variable before going to done.
|
2013-05-18 16:06:15 -07:00 |
|
William King
|
fc42d93ddf
|
FS-5441: --resolve
|
2013-05-17 12:40:04 -07:00 |
|
Anthony Minessale
|
c8ef74055a
|
FS-5430 --resolve
|
2013-05-15 09:27:44 -05:00 |
|
Anthony Minessale
|
e8ecb64666
|
FS-5410 --resolve
|
2013-05-13 13:15:54 -05:00 |
|
Raymond Chandler
|
d125147dd8
|
add context
|
2013-05-10 16:32:22 -04:00 |
|
Anthony Minessale
|
d6c2b9a2fb
|
FS-5085 the last change helped weed out the real bug which is this commit
|
2013-05-10 10:45:07 -05:00 |
|
Anthony Minessale
|
e5754ab3ee
|
FS-5403 --resolve
|
2013-05-10 09:59:19 -05:00 |
|
Anthony Minessale
|
3fc8f7888a
|
FS-5399 --resolve
|
2013-05-09 14:09:23 -05:00 |
|
Anthony Minessale
|
646b34a2a2
|
FS-5398
|
2013-05-09 10:10:06 -05:00 |
|
William King
|
569da261aa
|
FS-5395: False alarm on a reported valgrind memory leak. The io_private is alloc'd from a memory pool which valgrind sucks at tracking. No need to free this, because it'll get cleared when the memory pool on the rsession is destroyed.
|
2013-05-08 18:45:55 -07:00 |
|
William King
|
924e24a0de
|
FS-5395: --resolve Resolve another mod_rtmp crash where oh specific hangup scenarios there would be a free of variable that wasn't ever malloced
|
2013-05-08 16:53:15 -07:00 |
|
Anthony Minessale
|
2e19cbcc61
|
FS-5085 I discovered that the bowout is happening from both legs at once, we only need to check this condition on the A side of the loopback bridge
|
2013-05-08 13:17:45 -05:00 |
|
Brian West
|
7c8c758586
|
Undo previous change
|
2013-05-08 14:40:48 -05:00 |
|