4ba7e57683
Make an IsDisconnected interface and use it ( #1278 )
2022-12-31 12:53:02 +05:30
112d6fc18b
Reduced log verbosity for pieces that are stable ( #1274 )
2022-12-29 23:47:36 -08:00
86bf5cb62e
Ensure we create en Egress ID with PsRPC ( #1273 )
2022-12-30 13:46:45 +13:00
7778cdf2cd
Do not use the egress version stored in redis to decide whether to enable PsRPC. Use a conf entry instead ( #1262 )
2022-12-30 09:32:55 +13:00
5d3f644667
update psrpc ( #1266 )
2022-12-27 13:43:32 -08:00
988858a98a
Update dependencies to generic versions ( #1259 )
2022-12-26 22:29:13 -08:00
976d4ea9db
Update psrpc, egressStore interface ( #1256 )
...
* Update psrpc, egressStore interface
* psrpc v0.2.0
2022-12-24 00:49:31 -08:00
1a48cc6a8b
Track subscription operations per source track. ( #1248 )
2022-12-23 12:23:26 +05:30
6719a3c714
Updated egress rpc ( #1252 )
...
* updated egress rpc
* check if egress exists on stop
* fix static check
* remove old migration code
* rename
* regenerate, update test
* latest staticcheck
* update to psrpc 0.1.0
* fix tests
* dual write rpcs on running egress
* remove unused field
* fix race, change service for egress impl
* return nil if bus is nil
* id -> ids
* add affinityFunc to StartEgress
2022-12-22 21:03:27 -08:00
50e39b9985
Check participant SID also while removing a participant. ( #1237 )
2022-12-19 22:53:11 +05:30
120335da00
Allow skipping of sending ParticipantJoined analytics event ( #1236 )
...
In certain scenarios such as migration, we do not want a duplicate event
to be sent when the participant is reconnecting. The Prometheus metric
should still be updated though.
2022-12-18 22:09:20 -08:00
241a7120f5
ICE config using protocol model ( #1233 )
...
* ICE config using protocol model
* use pointers consistently
* protocol pointer
* mage generate
2022-12-19 10:25:08 +05:30
33902a9f2a
Do not send ParticipantLeft webhook event unless connected successfully. ( #1234 )
...
Fixes #1130
2022-12-18 17:37:55 -08:00
8a6c6de1db
update name of participant ( #1213 )
2022-12-15 22:03:59 -08:00
7a1273151f
Update to new logging library, using sampling participant logger ( #1219 )
2022-12-09 00:09:03 -08:00
6bd5504bff
Add option to issue full reconnect on a publication error. ( #1214 )
...
* Add option to issue full reconnect on a publication error.
Leaving the publication error timeout at 30 seconds as there
are some publications taking long. Also, there are cases
where the peer connection fails after 30 seconds. The peer
connection failure happens after publication error is detected.
But, 30 seconds is a good amount of time for publication to establish.
* prevent recursive lock
2022-12-06 14:46:59 +05:30
e9abb47020
Added logging fields for Ingress & Egress services ( #1205 )
2022-12-04 21:44:16 -08:00
14de2bec9c
Fixed single-node routing breakage. ( #1209 )
...
* Fixed single-node routing breakage.
Due to a regression of a previous change, Redis was always enabled even
when no configuration was provided.
* updated go modules
2022-12-04 16:23:35 -08:00
12ae179be2
Configurable RoomService execution timeout ( #1206 )
...
* API execution timeout is now configurable
In certain environments, it can take longer than the default 2s to
fully execute API requests. Making execution timeout a configurable option.
* do not expose api to YAML. internal for now.
2022-12-04 10:13:09 -08:00
d146ec7a1f
Improve logging messages with RoomService ( #1203 )
2022-11-30 22:17:28 -08:00
3c907ed460
Add stats for data channel and signal ( #1198 )
...
* Add stats for data channel and signal
* Solve comment
2022-11-30 14:53:19 +08:00
6711060cdb
Add enable loopback candidate option ( #1185 )
2022-11-23 16:01:36 +08:00
e5dabd466e
Support redis cluster mode ( #1181 )
...
* use redisConfig of protocol instead of redisConfig and use redis of protocol to create redis client to support redis cluster mode too
2022-11-22 10:36:43 -08:00
2c2c6f9da2
Do not append the stream key to the ingress URL for rtmp ( #1167 )
2022-11-15 09:19:25 -08:00
e2d775588f
Confirm room creation prior to returning from CreateRoom ( #1157 )
2022-11-09 23:47:41 -08:00
e5d21cb1d9
CreateRoom API to actually create the room on an RTC node ( #1155 )
...
Previously, CreateRoom only created the room in storage, but did not
hydrate it on an RTC node. This has caused strange behaviors such as
emptyTimeout not working correctly (#1109 ).
Also reduced room reap worker to consistently reap rooms. Fixes #241
2022-11-09 23:35:35 -08:00
7e89ad3fbd
RedisStore: make UnlockRoom atomic ( #1044 )
...
Co-authored-by: David Zhao <dz@livekit.io >
2022-11-07 23:04:46 -08:00
c735668f67
Use the redis.UniversalClient interface instead of *redis.Client when interacting with go-redis ( #1149 )
...
* Use the redis.UniversalClient interface instead of *redis.Client when interacting with go-redis
* Update protocol to v1.2.1
2022-11-07 17:27:28 -08:00
0310aa9250
Make sure client get participant info before track fired ( #1147 )
2022-11-07 14:50:45 +08:00
9a45b59414
Use ingress specific grants ( #1125 )
2022-10-26 21:37:36 -07:00
7223d9c132
web egress ( #1126 )
2022-10-26 13:43:56 -07:00
26fe910e88
Generated CLI Flags ( #1112 )
2022-10-25 22:24:08 -07:00
82c2ec8273
Remove named returns from room service. ( #1124 )
2022-10-26 10:24:13 +05:30
4344af6fd3
Some misc changes ( #1107 )
...
- ticker.Stop always
- clean up timer func (if they are added) on participant close
- sequencer test enhancement to add a real packet after a pdding packet
2022-10-20 11:11:45 +05:30
dff5379b78
remove record check on CreateRoom ( #1096 )
2022-10-17 11:25:20 -07:00
759e3bb1f2
Refine nat 1to1 mapping setting ( #1094 )
...
Now only set mapping when user_external_ip enabled or node_ip is
explicitly set. If multiple local address resolved to same external
ip, only the first one will be mapped to external, avoid candidate
conflict between different clients.
2022-10-17 16:11:52 +08:00
4161768530
Log Service API requests ( #1091 )
2022-10-17 00:16:54 -07:00
00ec859dd1
Add default handler to 404 ( #1088 )
2022-10-16 10:30:34 -07:00
78386c4669
Update protocol ( #1075 )
...
* update protocol
* auto egress ensure unique filename
* fix tests
2022-10-10 10:38:46 -07:00
1019faa0e6
Cleanup pass through logging ( #1073 )
...
* added filtering for noisy pion logs
* demoted some logs to debug
* using consistent trackID / participant / publisher / subscriber terminology
* removed ice candidate log lines, deferring to combined log
2022-10-06 23:48:37 -07:00
c3120679e6
Make sure CreateIngress returns an empty state ( #1069 )
2022-10-04 19:23:09 -07:00
8cf0872ec1
Adopt updated ingress RPC API ( #1068 )
2022-10-03 11:31:55 -07:00
716f27c435
Close UDPMux when exit ( #1064 )
...
* Close UDPMux when exit
* fix test failed
2022-10-01 09:03:58 +08:00
33f5dbc501
Make strict mode a param ( #1054 )
...
So, a wrapping config (like cloud) can skip the strict mode check.
2022-09-29 13:07:32 +05:30
932af81f34
Update strored version of an active ingress if no ingress server responds ( #1031 )
...
This allows deleting and updating an ingress even if the ingress server that was handling it died. It does however mean that if the ingress responds again later, its state will be inconsistent. To somewhat make this less likely, also keep trying contacting the ingress for 1 min in the background.
Also fixing a race where an active deleted Ingress would get recreated on delete because of the update triggered by the ingress session shutdown
2022-09-26 11:16:27 -07:00
5e1912e44c
Enable TCP/TURN fallback by default ( #1033 )
2022-09-22 23:58:07 -07:00
b97d59b8db
consolidate room internal ( #1030 )
...
* consolidate room internal
* create room internal map
* pipelined room read
* check error
* fix pipelined reads
* clean up after test
2022-09-22 15:59:27 -07:00
803046b882
Auto egress ( #1011 )
...
* auto egress
* fix room service test
* reuse StartTrackEgress
* add timestamp
* update prefixed filename explicitly
* update protocol
* clean up telemetry
* fix telemetry tests
* separate room internal storage
* auto participant egress
* remove custom template url
* fix internal key
* use map for stats workers
* remove sync.Map
* remove participant composite
2022-09-21 12:04:19 -07:00
feb47812e7
Allow CORS responses to be cached to allow faster initial connection ( #1027 )
2022-09-20 23:56:24 -07:00
99cb021e85
Increase max wait for media node to respond. ( #1026 )
...
In extreme cases, media nodes could take more than 5s to spin up the
session. Increasing this timeout to 10s reduces the number of disconnections
due to edge cases.
2022-09-20 23:38:58 -07:00
Raja Subramanian