github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 08:30:05 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[PR #1445] [MERGED] fix: parallel queueing race condition caused silent failure #15857

New Issue
Closed
opened 2026-04-16 05:10:28 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-16 05:10:28 -05:00
Owner
Copy Link

📋 Pull Request Information

Original PR: https://github.com/ollama/ollama/pull/1445
Author: @BruceMacD
Created: 12/9/2023
Status: ✅ Merged
Merged: 12/9/2023
Merged by: @jmorganca

Base: main ← Head: brucemacd/parallel

📝 Commits (2)

📊 Changes

1 file changed (+30 additions, -26 deletions)

View changed files

📝 llm/llama.go (+30 -26)

📄 Description

As of the most recent llama.cpp update concurrent requests had a race condition that would result in an empty response.

This was not easy to observe since the response from the llm runner subprocess was a 200 with the error {"content":"slot unavailable"} in the response stream, which just silently closed the channel.

This change resolves this by allowing multiple slots in the llm runner subprocess. We manage the queueing ourselves so this should be ok. @dhiltgen this may be a case we need to account for in the cgo changes.

🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.

## 📋 Pull Request Information **Original PR:** https://github.com/ollama/ollama/pull/1445 **Author:** [@BruceMacD](https://github.com/BruceMacD) **Created:** 12/9/2023 **Status:** ✅ Merged **Merged:** 12/9/2023 **Merged by:** [@jmorganca](https://github.com/jmorganca) **Base:** `main` ← **Head:** `brucemacd/parallel` --- ### 📝 Commits (2) - [`2439574`](https://github.com/ollama/ollama/commit/2439574b0fe04b4252dbcccf1c21a93a2bb3dcb8) fix: queued request failures - [`2a43e92`](https://github.com/ollama/ollama/commit/2a43e92698ba01fa31250335fcde64f384d98f07) log steam errors ### 📊 Changes **1 file changed** (+30 additions, -26 deletions) <details> <summary>View changed files</summary> 📝 `llm/llama.go` (+30 -26) </details> ### 📄 Description As of the most recent llama.cpp update concurrent requests had a race condition that would result in an empty response. This was not easy to observe since the response from the llm runner subprocess was a 200 with the error {"content":"slot unavailable"} in the response stream, which just silently closed the channel. This change resolves this by allowing multiple slots in the llm runner subprocess. We manage the queueing ourselves so this should be ok. @dhiltgen this may be a case we need to account for in the cgo changes. --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>
GiteaMirror added the pull-request label 2026-04-16 05:10:28 -05:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label pull-request
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#15857
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?