* Forwarding latency measurement tweaks.
- prom transmission type public
- do not measure short term values as it is not used and saves some lock
contention time in packet path potentially. Adding a separate method
for that.
- Change latency/jitter summary reporting to `ns` also to match the
histogram.
* add GetShortStats
* Higher resolution forwarding latency histogram.
Was using the average latency/jitter of last second to populate
forwarding latency/jitter histogram. But, it is too coarse, i. e. the
average value of latency/jitter is very low and those summarised samples
end up in the lowest bucket always.
A few things to address it
- record per packet forwarding latency in histogram
- adjust histogram bins to include smaller values
- Drop jitter histogram
This is a per packet call, but prometheus histogram is supposedly
fast/light weight. Would be good to get better resolution histograms.
Hence doing this. Please let me know if there are performance concerns.
* typo
* one more typo
* Add prom histogram for forwarding latency and jitter.
Using short term stats for histogram.
An example setting is
1s - short term
1m - long term
Using the 1s (short term) data for histogram. In that 1 second, all
packet forwarding latencies are averaged for latency and std. dev. of
the collection is used as jitter.
* try different staticcheck
Currently, the signal requests are counted on media side and signal
responses are counted on controller side. This does not provide the
granularity to check how many response messages each media node is
sending.
Seeing some cases where track subscriptions are slow under load. This
would be good to see if the media node is doing a lot of signal response
messages.
* Rework node stats a bit.
Related protocol PR - https://github.com/livekit/protocol/pull/1023
- Make a config for node stats measurements. Wanted to put the config in
`routing` package, but a circular dependency forced me to put in
config.go
- Make rate calculations explicit, i. e. requested via config.
Previously, it had some odd checks to decide when to calculate rate
and it would have been calculating over different windows.
- Report signal/data channel bytes every 5 seconds to stats collection
module. Previously, it was doing it every 30 seconds and that meant
some windows could have had a large spike
NOTE: Still need to think about this for load calculations as a large
number of participants leaving could flush in a small window and that
could report a large spike in bytes/packets. Maybe need to ignore
signal bytes for load calculation?
* deps
* use default node stats config if given config is nil
* split out node stats into a struct for re-use
* update config
* De-centralize some configs to where they are used.
And make default variables.
Renaming a bit, but these are all internal config and have not been
added to documented config.
* Keep documented config as is.
* test
* typo
* Add counter for pub&sub time metrics
The pub&sub shows large value in migration related case like
muted/disabled migration, the subscription time depends on
the time when publisher unmute the track(sending rtp packet
after migration), add a counter to distinguish since we
can't control the time in such cases and the first subscription
attemps also is more meaningful than those cases.
* Add info log for high publish delay
* Record out-of-packet count/rate in prom.
Adding a field to AnalyticsStream to make this easier to report.
Let me know if adding to AnalyticsStream is not ok.
Will set up a protocol PR if it is okay.
* deps
* speed up track publication
Add metrics for track publication and subscription
Return EnabledCodecs in JoinResponse so client can
choose codec without server side codec fallback
Cache remote webrtc track without AddTrackRequest to
let client send publisher offer before AddTrackRequest response
* go mod
* clean code
Because we aren't able to get CPU count/load info on Windows, they are
stubbed out to return placeholders. This restores compatibility to run
on Windows.
Sometimes the initial selected node could fail. In that case, we'll give it a few more attempts to locate a media node for the session instead of failing it after the first try.
Added a new manager to handle all subscription needs. Implemented using reconciler pattern. The goals are:
improve subscription resilience by separating desired state and current state
reduce complexity of synchronous processing
better detect failures with the ability to trigger full reconnect
* add prometheus stats for rtt/jitter/packet loss
* add track source to metrics
* better packet loss bins
* add track type to metrics
* remove source from AnalyticsStat
* regenerate telemetry service fake
* compute loss from per stream packet count
* initial commit
* add correct label
* clean up
* more cleanup on adding stats
* cleanup
* move things to pub and sub monitors, ensure stats are correctly updated
* fix merge conflict
* Fix panic on MacOS (#1296)
* fixing last feedback
Co-authored-by: Raja Subramanian <raja.gobi@tutanota.com>
