What's the chance "IoContext timed out due to inactivity" is caused by script upgrade?
Hoping someone with a better understanding of Durable Object RPC internals can help here. I'm seeing these two logs:
and in the same EventTimestampMs
I'm sending a ReadableStream over RPC from an AI workflow which is long running. Obviously when scripts upgrade (deployment or Workers platform decision) it terminates the RPC streams. My theories
- ReadableStream terminates due to script upgrade -> IoContext times out
- in other RPC calls (not long-running) we're also seeing this. It feels like stubs are kept open? My
and in the same EventTimestampMs
I'm sending a ReadableStream over RPC from an AI workflow which is long running. Obviously when scripts upgrade (deployment or Workers platform decision) it terminates the RPC streams. My theories
- ReadableStream terminates due to script upgrade -> IoContext times out
- in other RPC calls (not long-running) we're also seeing this. It feels like stubs are kept open? My
dispose
Welcome to the official Cloudflare Developers server. Here you can ask for help and stay updated with the latest news
85,042Members
Resources
Similar Threads
Was this page helpful?
