Stops waiting for ollama before specified timeout reached #7635
Description
Describe the bug
When trying to use Ollama with any decent model I run into error:
"Ran into this error: Server error: Server error (500 Internal Server Error): health resp: Get "http://127.0.0.1:60954/health": dial tcp 127.0.0.1:60954: connectex: No connection could be made because the target machine actively refused it..
Please retry if you think this is a transient or recoverable error."
It seems to wait Ollama for 5-10 seconds to respond and gives up.
This does not happen if i use very small model such as 2b parameter variant of Qwen3.5.
With small model it patiently waits for model to respond even over minute.
💡 Before filing, please check common issues:
https://block.github.io/goose/docs/troubleshooting
📦 To help us debug faster, attach your diagnostics zip if possible.
👉 How to capture it: https://block.github.io/goose/docs/troubleshooting/diagnostics-and-reporting/
A clear and concise description of what the bug is.
To Reproduce
Steps to reproduce the behavior:
- Use local Ollama
- Select model that takes while to respond. In my case with RTX3060 12GB and 64GB system ram Qwen3.5 35b model that still works fine in Ollama but takes 10-15 seconds to start responding.
- Ask it anything
- See the error in chat
Expected behavior
Get response
Please provide the following information
- OS & Arch: Windows 11
- Interface: UI
- Version: 1.26.1
- Extensions enabled: Apps, Code mode, Computer Controller, Developer, Extension Manger, Memory, Summon, Todo, Top Of Mind. But note with these work fine on smaller model
- Provider & Model: Local Ollama qwen3.5:35b
Additional context
Qwen3.5:35b from my local Ollama works fine on Ollama terminal and for other tools that use Ollama as backend.