LiveKit Community

Also looks like connect waits for ICE now which is great, can trust it. probably need a timeout there

i didn’t update the front end library though (should i?)

are you using this commit on client?

```commit 0dbf095f358d3d3d8183cda4a0158f1fbfc3debe (HEAD -&gt; main, origin/main, origin/HEAD)
Author: David Zhao &lt;david@davidzhao.com&gt;
Date:   Sat Jan 30 14:18:00 2021 -0800

    rename Participant.name to Participant.identity```

Don’t think so, is there an updated client version or I should build locally

I'll push a version.. that's probably why

just pulled and looked at the changes, makes sense! lmk when you push a version

I forgot the clients are speaking JSON instead of protobufs to the server, so it's sensitive to field name changes

fixed the identity “issue” i was seeing, thanks!

its empty on the publishing localParticipant too

need to add a test case there.. I'll have a look

In general don’t like identity how it is b/c we need to atob + JSON.parse + get array index 3 to get the userId value

i think if it was an id field or a map that i can set values for

does it have to be a json encoded string?

Screen Shot 2021-01-31 at 1.15.29 PM.png

here’s the other side keeping track of users tracks + making a mediastream

I don’t think we set participant.identity, it comes back to us in this way

identity is the string you encoded in the access token

it’s possible user.id isn’t the actual id

I'm not following. LK would pass this string to you. it's opaque to livekit

Ok it’s in my code where the id is encoded

Screen Shot 2021-01-31 at 2.07.11 PM.png

ok <@U01KZ61GV5Z> next one for you :slightly_smiling_face:. This happens on refresh sometimes

on a page connected to backend -&gt; refresh -&gt; logs show this when trying to connect

does the server logs state the same as what you reported above? i.e. room already closed?

ok, rolled out 0.3.1, that no longer close rooms as aggressively after participant disconnect