Closed artur9010 closed 4 months ago
When we upload we try all hosts. It shouldn't cause uploads to fail. However, I'll add this to the backlog since we might want to change the code slightly to not bother hosts that ran out of storage too much.
@ChrisSchinnerl upload penalty should de-prioritise hosts with full disks. Not sure if we need to do more than that. The issue described here is likely due to a lack of contracts in general and just testnet hosts being testnet hosts.
Closing this for now since we PJ is right and we confirmed that with additional testing around host penalties recently.
Current Behavior
autopilot is trying to migrate all of slabs to host with full disk
response from host:
Expected Behavior
x
Steps to Reproduce
No response
Version
402398b, zen
What operating system did the problem occur on (e.g. Ubuntu 22.04, macOS 12.0, Windows 11)?
Kubernetes, https://github.com/artur9010/charts/tree/master/renterd
Autopilot Config
probably not necessary
Bus Config
probably not necessary
Contract Set Contracts
probably not necessary
Anything else?
No response