Richard Fuchs
|
fc246e64d3
|
cant guarantee that the call struct is still valid when the response is sent, to copy strings
|
12 years ago |
|
Richard Fuchs
|
0230985e91
|
get rid of another useless variable
|
12 years ago |
|
Richard Fuchs
|
55380807f8
|
tags can be NULL
|
12 years ago |
|
Richard Fuchs
|
bf4df4dff9
|
remove more code duplication
|
12 years ago |
|
Richard Fuchs
|
504995e47c
|
return call stats in NG delete message reply
|
12 years ago |
|
Richard Fuchs
|
feb3b6bde1
|
adding some const for good measure
|
12 years ago |
|
Richard Fuchs
|
3c764bed52
|
split NG stats collection into ng_call_stats()
|
12 years ago |
|
Richard Fuchs
|
f7da33b4a7
|
adding a UINT64F format macro
|
12 years ago |
|
Richard Fuchs
|
b2ad672086
|
make call_query_udp use the new call stats function
|
12 years ago |
|
Richard Fuchs
|
495d4cf2bd
|
keep track of "newest" timestamp
|
12 years ago |
|
Richard Fuchs
|
267252542b
|
create a struct for stats collection
|
12 years ago |
|
Richard Fuchs
|
1adef8fe7b
|
split stream traversal for stats query into separate function
|
12 years ago |
|
Richard Fuchs
|
8c9fa012bd
|
lets have a totals packet/bytes counter for each call
|
12 years ago |
|
Richard Fuchs
|
e9bf80768d
|
add more bencode convenience functions
|
12 years ago |
|
Richard Fuchs
|
1ef1c87545
|
add a prelim NG query command
|
12 years ago |
|
Richard Fuchs
|
aaa3f4d116
|
simplify call_query_udp a bit
|
12 years ago |
|
Richard Fuchs
|
f369c9f799
|
more redis support functions
|
12 years ago |
|
Richard Fuchs
|
8d3b8e310a
|
redis support functions
|
12 years ago |
|
Richard Fuchs
|
20f33e56ff
|
treat zero ip/port as wildcard when matching up old streams
|
12 years ago |
|
Richard Fuchs
|
5cebf5adbd
|
use peer_advertised where appropriate
|
12 years ago |
|
Richard Fuchs
|
fc6fe69075
|
replace IN6_IS_ADDR_UNSPECIFIED because it doesnt work for mapped addresses
|
12 years ago |
|
Richard Fuchs
|
5d726fdf84
|
fix some ipv6 related bugs
|
12 years ago |
|
Richard Fuchs
|
6adda7c301
|
keep track of sdp stream endpoints and use them as index when rewriting
|
12 years ago |
|
Richard Fuchs
|
bac4f31489
|
get rid of num parameter to sdp_replace
|
12 years ago |
|
Richard Fuchs
|
7ca74b399e
|
use iovec based sdp rewriting
|
12 years ago |
|
Richard Fuchs
|
df1b169cba
|
handle consecutive ports in SDP
|
12 years ago |
|
Richard Fuchs
|
a67f03d1d3
|
introduce struct udp_fd
|
12 years ago |
|
Richard Fuchs
|
4c14016f9f
|
code cosmetics
|
12 years ago |
|
Richard Fuchs
|
7137b784e9
|
implement a "relays cache"
|
12 years ago |
|
Richard Fuchs
|
061507585c
|
variable renaming - getting confusing
|
12 years ago |
|
Richard Fuchs
|
a45a7379ed
|
obsolete get_port_pair()
|
12 years ago |
|
Richard Fuchs
|
ce2df6d90f
|
exclude our own listening ports from RTP
|
12 years ago |
|
Richard Fuchs
|
9ae844040a
|
reimplement get_port_pair() to support multiple consecutive ports
|
12 years ago |
|
Richard Fuchs
|
e22b6a04fb
|
support trust-address flag
|
12 years ago |
|
Richard Fuchs
|
ab3a47ae41
|
finish renaming control.[ch] to control_tcp.[ch]
|
12 years ago |
|
Richard Fuchs
|
efdc8efeb8
|
rename some command implementation functions for consistency
|
12 years ago |
|
Richard Fuchs
|
9f78c2f17a
|
implement call_delete_ng()
|
12 years ago |
|
Richard Fuchs
|
3069720052
|
fix copy-pasto
|
12 years ago |
|
Richard Fuchs
|
2872d18f43
|
implement call_answer
|
12 years ago |
|
Richard Fuchs
|
5d1c2aabed
|
support some flags passed from rtpproxy-ng module
|
12 years ago |
|
Richard Fuchs
|
5accfd0751
|
dont return ipv6 address by default
|
12 years ago |
|
Richard Fuchs
|
011e0e7e4e
|
unify call_update_udp and call_lookup_udp
|
12 years ago |
|
Richard Fuchs
|
e5d7341bdc
|
unify call_request and call_lookup
|
12 years ago |
|
Richard Fuchs
|
6fd39168f1
|
obsolete infohash
|
12 years ago |
|
Richard Fuchs
|
ecc37e4b2a
|
fix segfault related to null log_info
|
12 years ago |
|
Richard Fuchs
|
b66809bb8c
|
use new struct stream_input where appropriate
|
12 years ago |
|
Richard Fuchs
|
8fe26b2d3b
|
obsolete the "mediatype" element
|
12 years ago |
|
Richard Fuchs
|
9589a73b42
|
obsolete call{ed,ing}_agent element
|
12 years ago |
|
Richard Fuchs
|
337167faf7
|
use enum stream_address_format
|
12 years ago |
|
Richard Fuchs
|
923a7bdf5a
|
correctly mark ip6 address
|
12 years ago |
|
Richard Fuchs
|
c75394d3cd
|
first take on SDP rewriting
|
12 years ago |
|
Richard Fuchs
|
2b60a82e36
|
more mass deployment of the new "str" object - does this thing still work? now really?
|
12 years ago |
|
Richard Fuchs
|
60da1bc445
|
mass deployment of the new "str" object - does this thing still work?
|
12 years ago |
|
Richard Fuchs
|
8db7ab6c94
|
introduce kamailio-style "str" data type/struct
also remove several occurences of "const" that keep biting me in the rear.
|
12 years ago |
|
Richard Fuchs
|
ccfa304354
|
sdp_streams() initial implementation
|
12 years ago |
|
Richard Fuchs
|
f6a336ce89
|
prefer objects on stack instead of heap when possible
|
12 years ago |
|
Richard Fuchs
|
a2ac8075a9
|
WIP on SDP parser
|
12 years ago |
|
Richard Fuchs
|
be4a6c7d58
|
make my life easier with a smart_ntop_port() function
|
12 years ago |
|
Richard Fuchs
|
5ea5dcddfa
|
call_offer()/answer stubs
|
12 years ago |
|
Richard Fuchs
|
bca25d7700
|
make call_lookup_udp() remember viabranches like call_update_udp()
|
12 years ago |
|
Richard Fuchs
|
ef521030ab
|
make viabranch in udp lookup optional
|
12 years ago |
|
Richard Fuchs
|
ed7f8dda04
|
remove obsolete code; add more locking
|
13 years ago |
|
Richard Fuchs
|
644e5ba9df
|
calculate proper "TTL"
|
13 years ago |
|
Richard Fuchs
|
bbb8327750
|
implement udp Q command
|
13 years ago |
|
Richard Fuchs
|
a99ada4e41
|
add more logging to xmlrpc call mechanism
|
13 years ago |
|
Richard Fuchs
|
eda487e579
|
un-shoot myself in the foot, gna
|
13 years ago |
|
Richard Fuchs
|
8092a56761
|
another locking fix
|
13 years ago |
|
Richard Fuchs
|
bb054ec00b
|
fix more locking problems
|
13 years ago |
|
Richard Fuchs
|
3a26e9fceb
|
fix possible race condition
|
13 years ago |
|
Richard Fuchs
|
39207a0859
|
dont fail on zero port in tcp controller
|
13 years ago |
|
Richard Fuchs
|
37013fd148
|
IPv6 socket requires IPv4 PKTINFO for setting source address correctly
|
13 years ago |
|
Richard Fuchs
|
59c5ffd0da
|
fix reference count leak
|
13 years ago |
|
Richard Fuchs
|
08d40d8ac5
|
add debian buildflags and address generated warnings
|
13 years ago |
|
Richard Fuchs
|
594297a3ea
|
use string chunk for all the little call-related strings
|
13 years ago |
|
Richard Fuchs
|
afb1f5de95
|
need to lock the call a bit earlier
|
13 years ago |
|
Richard Fuchs
|
d8ef1885cc
|
split callstream_init into callstream_new for redis module
|
13 years ago |
|
Richard Fuchs
|
c291d1220b
|
fix logging inconsistency
|
13 years ago |
|
Richard Fuchs
|
72840c8d1d
|
reorder calls to redis for proper locking
|
13 years ago |
|
Richard Fuchs
|
d20d747a63
|
since xmlrpc is still segfaulting even in its own process, disable core dumps in the child and retry the xmlrpc call up to 3 times if the child terminates abnormally. jumping through hoops ftw!
|
13 years ago |
|
Richard Fuchs
|
2ad1d6f65e
|
xmlrpc still segfaulting. leaves no choice but to fork out a new process, ugh.
|
13 years ago |
|
Richard Fuchs
|
f4833f6b74
|
xmlrpc calls still causing segfaults, lets try with a dedicated client object per call... i'm starting to dislike this library
|
13 years ago |
|
Richard Fuchs
|
fabcb53a3b
|
the curl xmlrpc transport isnt playing nicely - revert to default
|
13 years ago |
|
Richard Fuchs
|
08c764c748
|
cant decref null references
|
13 years ago |
|
Richard Fuchs
|
0f6d898171
|
add proper locking to callstream setup procedure
|
13 years ago |
|
Richard Fuchs
|
94f2f320a6
|
actually the call struct needs neither linked flag nor shutdown flag due to the sequence of events and locking
|
13 years ago |
|
Richard Fuchs
|
d8776907f8
|
need to lock the callstream a bit earlier
|
13 years ago |
|
Richard Fuchs
|
ab3b8c60f0
|
fix a silly, threading-unrelated, statistics-skewing bug
|
13 years ago |
|
Richard Fuchs
|
45f34b9784
|
turns out you actually have to free/destroy mutexes. who woulda thunk?
|
13 years ago |
|
Richard Fuchs
|
54353022ff
|
om nom nom memory leak... well, reference count leak
|
13 years ago |
|
Richard Fuchs
|
b7a68659b1
|
analoguous to the last race, call_destroy() must not assume that it's getting the last reference to the call
|
13 years ago |
|
Richard Fuchs
|
ddffe5eeec
|
global xmlrpc init must be done prior to any threading
|
13 years ago |
|
Richard Fuchs
|
d1fbf9bef9
|
alarms arent usable in a threaded environment, so use curl internal timeout instead
|
13 years ago |
|
Richard Fuchs
|
feeeee2fab
|
must extract needed tags for xmlrpc call teardown prior to destroying the calls
|
13 years ago |
|
Richard Fuchs
|
973d92d9c7
|
moving call teardown due to timeout to a separate thread
|
13 years ago |
|
Richard Fuchs
|
2dd6cfe68c
|
keeping the lock hold time to a minimum in lengthy calls_status operation
|
13 years ago |
|
Richard Fuchs
|
8c2c69f5e6
|
make the stats collection thread-safe
|
13 years ago |
|
Richard Fuchs
|
d9e09f33bc
|
lets try with a lot of large, nested, naive locks
|
13 years ago |
|
Richard Fuchs
|
ae5a36f5ce
|
getting rid of call->log_info and introducing call->lock
|
13 years ago |
|
Richard Fuchs
|
900f76a17e
|
make the callmaster struct private - also requires update to redis module
|
13 years ago |
|
Richard Fuchs
|
aeb5398b0d
|
granular lock for "ports used" bit field
|
13 years ago |
