What causes this? This was one conversation, now i...
# general
c
What causes this? This was one conversation, now it's split into a few and I'm getting duplicate messages?
m
Interesting. Could you share the versions you are using? This happens in a particular situation or just randomly?
c
1.18.0
First time I've seen it do this, but ever version I've used slows down a lot and typing into the message box is a pain. Then it will loose connection to the back end and reconnect and it's all fine again.
m
Cool thanks… i will give a look tomorrow
c
Here's what I was talking about. After sometime of using OH, especially long projects, I'll notice the very slow message typing and it will get worse until I'll get this screen. A browser refresh gets me back in business and now the tying is fine for awhile until it starts again.
On the VM/Host side, everything looks fine.
m
I had finally the time to look into that. I think there are a couple of issues here. 2 are actually already documented: • Conversation UI can enter a transient React/DOM rendering storm and become unresponsive • [Bug]: Duplicate messages from the agent in 1.12.0 While 1 is new: • [Bug]: WebSocket reconnect resubscribes from a stale `after_timestamp`, replaying the whole session
for the issue 2: we ar reworking how stream works. So probably this would be solved soon
c
Constantly get this message with deep seek 4
Curious, do you guys us your own product? Seems like us average users find all kinds of bugs that should have been found by normal qa. Not trying to be rude just curious how much the team uses it's own tools.
Or, am I wrong in assuming OH is ready for production usage. Maybe my expectations are too high. If that's the case ok but I'm tired of wasting money and time on what might be product testing.