Sorry about the format of this report .. this is in a Linux Mint environment ... but probably affects others.
I sometimes get network errors like this:
Based on the image content you provided, here is the exact text:
reasoning done · 57s
The user wants me to investigate how hard/easy it would be to pin down the "content half" – the distinction between "natural language (translatable)" vs "machine value (enum/status/code, etc.)"
[interrupted] Let me investigate the actual string data to ground the feasibility assessment rather than speculate.
Warn Provider stream connection dropped while reading the response body. Some output had already streamed, so Codewhale is surfacing the warning instead of replaying the request and risking duplicated output. Details: Stream read error: error decoding response body
Alt+V:full error
Warn Network error: SSE stream request failed after HTTP/1.1 fallback: error sending request for url (https://api.deepseek.com/v1/chat/completions). codewhale doctor can still pass when non-streaming requests work; on Windows or proxy networks, try CODEWHALE_FORCE_HTTP1=1 and rerun codewhale.
Alt+V:full error
network had an error - try to resume what you were doing
: working · 41s · Esc to interrupt $1.10 | 11 turns | LLM 30m13s | Cache hit 100% | Input 12.2M
===========================================================================
The problem is not the network error - usually things recover when the network returns
But sometimes, this followed by another error suggesting the codewhale engine has stopped (not the UI obviously). I have not managed to capture that yet. I will add it to this bug when it happens again.
The workaound is to exit codewhale, and then 'codewhale --continue' ... and hope things recover - they usually do
Sorry about the format of this report .. this is in a Linux Mint environment ... but probably affects others.
I sometimes get network errors like this:
Based on the image content you provided, here is the exact text:
reasoning done · 57s
The user wants me to investigate how hard/easy it would be to pin down the "content half" – the distinction between "natural language (translatable)" vs "machine value (enum/status/code, etc.)"
[interrupted] Let me investigate the actual string data to ground the feasibility assessment rather than speculate.
Warn Provider stream connection dropped while reading the response body. Some output had already streamed, so Codewhale is surfacing the warning instead of replaying the request and risking duplicated output. Details: Stream read error: error decoding response body
Alt+V:full error
Warn Network error: SSE stream request failed after HTTP/1.1 fallback: error sending request for url (https://api.deepseek.com/v1/chat/completions).
codewhale doctorcan still pass when non-streaming requests work; on Windows or proxy networks, tryCODEWHALE_FORCE_HTTP1=1and reruncodewhale.Alt+V:full error
network had an error - try to resume what you were doing
: working · 41s · Esc to interrupt $1.10 | 11 turns | LLM 30m13s | Cache hit 100% | Input 12.2M
===========================================================================
The problem is not the network error - usually things recover when the network returns
But sometimes, this followed by another error suggesting the codewhale engine has stopped (not the UI obviously). I have not managed to capture that yet. I will add it to this bug when it happens again.
The workaound is to exit codewhale, and then 'codewhale --continue' ... and hope things recover - they usually do