I’ve set up a scalable Jitsi Meet instance corresponding to this guide and so far the load balancing seems to be working fine. One thing I was wondering about however is the case of a single bridge failure.
When manually stopping one bridge with an active conference on it, the conference will instantly be moved to another bridge, but stream errors start to appear and the jvb.log is flooded with these:
2020-06-06 20:18:14.154 WARNING:  [confId=6e5669b84eb54d3d epId=c51f51b4 gid=ff06ab stats_id=Corrine-5v3 conf_name=testertest] TransportCcEngine.tccReceived#157: TCC packet contained received sequence numbers: 9310-9426. Couldn't find packet detail for the seq nums: 9310-9426. Latest seqNum was 4789, size is 1000. Latest RTT is 26.025948 ms.
This will continue until the conference ends and is disposed of by the server. Is it possible that I’m missing some configuration thing that could prevent this? Or is it expected behavior?