github-starred/ollama-ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-04-27 16:36:54 -05:00
Code Issues 2.2k Packages Projects Releases 100 Wiki Activity

Ollama hangs in infinite loop during code update requests, requires service restart #8649

New Issue
Open
opened 2025-11-12 14:48:14 -06:00 by GiteaMirror · 2 comments
GiteaMirror commented 2025-11-12 14:48:14 -06:00
Owner
Copy Link

Originally created by @A on GitHub (Nov 10, 2025).

What is the issue?

When requesting code updates through various client tools, Ollama enters an infinite loop state where:

  • GPU utilization remains active (visible in nvtop)
  • No output is produced to the client (in rare cases, it loops one block of text over and over)
  • The process continues even after canceling the task in the client
  • ollama stop command has no effect
  • Only systemctl restart ollama resolves the issue

Models tested (with different context size up to 32k):

  • maryasov/qwen2.5-coder-cline:32b
  • llama3.1:70b
  • codellama:34b-code

Clients tested:

  • avante.nvim
  • Cline for VSCode
  • Roocode

Description:

OS

Linux ubuntu 6.8.0-59-generic #61~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Tue Apr 15 17:03:15 UTC 2 x86_64 x86_64 x86_64 GNU/Linux

GPU

  • Nvidia RTX A4000
  • 2x RTX 3090 on different vast.ai servers

CPU

Not important

Ollama version

0.12.10

Originally created by @A on GitHub (Nov 10, 2025). ### What is the issue? When requesting code updates through various client tools, Ollama enters an infinite loop state where: - GPU utilization remains active (visible in nvtop) - No output is produced to the client (in rare cases, it loops one block of text over and over) - The process continues even after canceling the task in the client - ollama stop <model> command has no effect - Only systemctl restart ollama resolves the issue Models tested (with different context size up to 32k): - maryasov/qwen2.5-coder-cline:32b - llama3.1:70b - codellama:34b-code Clients tested: - avante.nvim - Cline for VSCode - Roocode Description: ```shell ``` ### OS Linux ubuntu 6.8.0-59-generic #61~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Tue Apr 15 17:03:15 UTC 2 x86_64 x86_64 x86_64 GNU/Linux ### GPU - Nvidia RTX A4000 - 2x RTX 3090 on different vast.ai servers ### CPU Not important ### Ollama version 0.12.10
GiteaMirror added the bug label 2025-11-12 14:48:14 -06:00
GiteaMirror commented 2025-11-12 14:48:15 -06:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 11, 2025):

Ollama looping is usually due to the model losing track of what it was doing and generating a stream of tokens that doesn't include an end-of-sequence token. This can be triggered by overflowing the context buffer, causing either message truncation at the time of prompt processing, or causing a context shift during token generation. If you could add a log with OLLAMA_DEBUG=1 set that would help in debugging.

@rick-github commented on GitHub (Nov 11, 2025): Ollama looping is usually due to the model losing track of what it was doing and generating a stream of tokens that doesn't include an end-of-sequence token. This can be triggered by overflowing the context buffer, causing either message truncation at the time of prompt processing, or causing a context shift during token generation. If you could add a log with `OLLAMA_DEBUG=1` set that would help in debugging.
GiteaMirror commented 2025-11-12 14:48:15 -06:00
Author
Owner
Copy Link

@pdevine commented on GitHub (Nov 12, 2025):

@A are you hitting this when you've run through the context, or some other case?

@pdevine commented on GitHub (Nov 12, 2025): @A are you hitting this when you've run through the context, or some other case?
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
client2
cloud
compatibility
context-length
create
docker
documentation
embeddings
engine
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
install
integration
intel
js
linux
macos
memory
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
windows
wsl
No Label bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama-ollama#8649
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?