github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-09 01:59:50 -05:00
Code Issues 14.7k Packages Projects Releases 122 Wiki Activity

[PR #6844] [CLOSED] runner.go: Don't panic when processing sequences #12241

New Issue
Closed
opened 2026-04-12 23:52:44 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-12 23:52:44 -05:00
Owner
Copy Link

📋 Pull Request Information

Original PR: https://github.com/ollama/ollama/pull/6844
Author: @jessegross
Created: 9/17/2024
Status: ❌ Closed

Base: jmorganca/llama ← Head: jessegross/no_panic

📝 Commits (4)

  • ac003ca runner.go: Update TODOs
  • 1547609 runner.go: Don't panic when processing sequences
  • 5e021a4 runner.go: Remove stop tokens from cache
  • 94a48ab runner.go: Allocate batches for all sequences during init

📊 Changes

3 files changed (+98 additions, -62 deletions)

View changed files

📝 llama/runner/cache.go (+11 -7)
📝 llama/runner/cache_test.go (+4 -2)
📝 llama/runner/runner.go (+83 -53)

📄 Description

If there is an error processing a sequence, we should simply return
an HTTP error and abort that sequence rather than panic the whole
runner. This will make us more resilient to transient failures.

Panics can still occur during startup as there is no way to serve
requests if that fails.

Based on some code that was originally part of the vision work.

Co-authored-by: jmorganca jmorganca@gmail.com

🔄 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/6844 **Author:** [@jessegross](https://github.com/jessegross) **Created:** 9/17/2024 **Status:** ❌ Closed **Base:** `jmorganca/llama` ← **Head:** `jessegross/no_panic` --- ### 📝 Commits (4) - [`ac003ca`](https://github.com/ollama/ollama/commit/ac003ca715bc7b89066012dce7b88035fd3d61e7) runner.go: Update TODOs - [`1547609`](https://github.com/ollama/ollama/commit/1547609b46b9bb23a217f1bbbb1bec917ed05a2d) runner.go: Don't panic when processing sequences - [`5e021a4`](https://github.com/ollama/ollama/commit/5e021a417135683677e26290d8d37dfddd79622a) runner.go: Remove stop tokens from cache - [`94a48ab`](https://github.com/ollama/ollama/commit/94a48ab1b3ccdb3330d1e40ef6a1ad9b0f3710fb) runner.go: Allocate batches for all sequences during init ### 📊 Changes **3 files changed** (+98 additions, -62 deletions) <details> <summary>View changed files</summary> 📝 `llama/runner/cache.go` (+11 -7) 📝 `llama/runner/cache_test.go` (+4 -2) 📝 `llama/runner/runner.go` (+83 -53) </details> ### 📄 Description If there is an error processing a sequence, we should simply return an HTTP error and abort that sequence rather than panic the whole runner. This will make us more resilient to transient failures. Panics can still occur during startup as there is no way to serve requests if that fails. Based on some code that was originally part of the vision work. Co-authored-by: jmorganca <jmorganca@gmail.com> --- <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 2026-04-12 23:52:44 -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#12241
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?