github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-05 23:53:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #4579] Redownloading model on run command after runner crash #28633

New Issue
Closed
opened 2026-04-22 07:06:17 -05:00 by GiteaMirror · 3 comments
GiteaMirror commented 2026-04-22 07:06:17 -05:00
Owner
Copy Link

Originally created by @TipuatGit on GitHub (May 22, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/4579

Originally assigned to: @dhiltgen on GitHub.

What is the issue?

I downloaded llama3 with the ollama run llama3 command, and after it downloaded, first it failed with the error:
Error: llama runner process has terminated: exit status 0xc0000005

Then to see if the error reproduces, I ran ollama run llama3 but it started to download it all over again. I don't understand why its doing that. And its repeatedly doing that, everytime I run the command it just goes to redownloading.

Note: I changed model directory by creating environment variable OLLAMA_MODELS as per the instructions in F.A.Qs. Also, model is in both the C drive and my other drive that I chose.

Foremost, I would like it to stop redownloading and use what is on my system already. That is top priority.
Please help guys.

OS

Windows

GPU

Intel

CPU

Intel

Ollama version

0.1.38

Originally created by @TipuatGit on GitHub (May 22, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/4579 Originally assigned to: @dhiltgen on GitHub. ### What is the issue? I downloaded llama3 with the `ollama run llama3` command, and after it downloaded, first it failed with the error: `Error: llama runner process has terminated: exit status 0xc0000005` Then to see if the error reproduces, I ran `ollama run llama3` but it started to download it all over again. I don't understand why its doing that. And its repeatedly doing that, everytime I run the command it just goes to redownloading. Note: I changed model directory by creating environment variable `OLLAMA_MODELS` as per the instructions in F.A.Qs. Also, model is in both the C drive and my other drive that I chose. Foremost, I would like it to stop redownloading and use what is on my system already. That is top priority. Please help guys. ### OS Windows ### GPU Intel ### CPU Intel ### Ollama version 0.1.38
GiteaMirror added the needs more infobugwindows labels 2026-04-22 07:06:17 -05:00
GiteaMirror commented 2026-04-22 07:06:20 -05:00
Author
Owner
Copy Link

@dp289m commented on GitHub (May 27, 2024):

Same here.
Intel i9 Macbook Darwin 23.1.0
ollama 0.1.38

<!-- gh-comment-id:2132675757 --> @dp289m commented on GitHub (May 27, 2024): Same here. Intel i9 Macbook Darwin 23.1.0 ollama 0.1.38
GiteaMirror commented 2026-04-22 07:06:20 -05:00
Author
Owner
Copy Link

@zombodotcom commented on GitHub (Jun 29, 2024):

Im experiencing this issue from https://github.com/ollama/ollama/issues/2551#issuecomment-2198366862
#2551

Changed environment variables cause even after the command line install and specifying directory, still downloaded to c drive

<!-- gh-comment-id:2198368198 --> @zombodotcom commented on GitHub (Jun 29, 2024): Im experiencing this issue from https://github.com/ollama/ollama/issues/2551#issuecomment-2198366862 #2551 Changed environment variables cause even after the command line install and specifying directory, still downloaded to c drive
GiteaMirror commented 2026-04-22 07:06:21 -05:00
Author
Owner
Copy Link

@dhiltgen commented on GitHub (Oct 23, 2024):

Looks like this issue slipped through the cracks.

We don't currently support Intel GPUs, so I presume you're running on CPU. We should try to understand the crash, so the first thing I'd recommend is upgrade to the latest version, and if you're still seeing a crash, please share the server log so we can see why it crashed.

Occasionally we update models with fixes for templates, etc., and when you perform ollama pull we'll check to see if it has changed and pull only if it did. ollama run should not do this, and will use the existing pulled model even if it has changed on ollama.com. It's possible there's some filesystem corruption, running out of disk space, or other failure that's causing the model to be corrupted or removed and leading to the re-pull. Again, server logs will help diagnose.

<!-- gh-comment-id:2433467985 --> @dhiltgen commented on GitHub (Oct 23, 2024): Looks like this issue slipped through the cracks. We don't currently support Intel GPUs, so I presume you're running on CPU. We should try to understand the crash, so the first thing I'd recommend is upgrade to the latest version, and if you're still seeing a crash, please share the server log so we can see why it crashed. Occasionally we update models with fixes for templates, etc., and when you perform `ollama pull` we'll check to see if it has changed and pull only if it did. `ollama run` should not do this, and will use the existing pulled model even if it has changed on ollama.com. It's possible there's some filesystem corruption, running out of disk space, or other failure that's causing the model to be corrupted or removed and leading to the re-pull. Again, server logs will help diagnose.
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 bug needs more info windows
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#28633
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?