Apologies don’t mean to be rude but “hoping” is not a solution here. I would appreciate some clarify if you can help debug the Room IDs shared above.
Happy to raise a support ticket if that is what is needed here to get more clarify - just lmk
Agreed, testing will be out best bet here.
It is still not clear to me why the agent pod’s WS connect to the LiveKit edge stall past 5s, when the region-URL retry moments later succeeds?
Have you tried any DNS caching to see if that helps?
The added logging in the newer version may help isolate the issue further.
I think the current confusion to me is - the problem itself.
- What is this issue we are seeing?
- Is it a client side issue (agents sdk connecting issue) or livekit cloud networking issue?
- Why is it occuring only to us since we have been the once reporting it?
Can you help with 1 and 2 for me to answer and dig deeper on 3 
- What is this issue we are seeing?
As I understand it you are seeing an error in your agent logs that I do not have access to. I have the RM_* ids and looking in the server logs but the key details are in the agent/k8s logs/env
- Is it a client side issue (agents sdk connecting issue) or livekit cloud networking issue?
To me it looking like a connectivity or TLS/DNS issue. I can’t really tell from our side on this one.
- Why is it occuring only to us since we have been the once reporting it?
I have the same questions. Any “V0” issues I see folks have reported are old and seem to no longer occur.
There is some added logging that will let us dig a little deeper into what state the code makes it too once you are running it to pick up the other fixes.
which version of agents sdk were these logs added in?
The next agent version that is in progress of being released today.
Looking at the code
It now surfaces the full error chain on region‑fetch failures (so a TLS/DNS/connection failure on the fallback shows the real cause instead of a generic “error sending request”) in livekit-api v0.5.2 (Jun 17 2026)