github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 16:40:08 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #932] Failed to parse available VRAM: strconv.ParseInt: parsing "[Insufficient Permissions]" #46969

New Issue
Closed
opened 2026-04-28 02:19:12 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-28 02:19:12 -05:00
Owner
Copy Link

Originally created by @domWinter on GitHub (Oct 27, 2023).
Original GitHub issue: https://github.com/ollama/ollama/issues/932

Originally assigned to: @BruceMacD on GitHub.

Hi,

I am trying to run ollama in a docker container with the nvidia runtime on a system with a Nvidia A100 and MIG enabled.
When starting the ollama server I get the following error:

2023/10/27 09:16:05 routes.go:682: Warning: GPU support may not enabled, check you have installed install GPU drivers: failed to parse available VRAM: strconv.ParseInt: parsing "[Insufficient Permissions]": invalid syntax

The ollama server is running as root user.

However, when running nvidia-smi inside the container I see the passed through GPU and MIG device and I am also able to allocate the GPU with e.g. pytorch in the same container.

Cuda Version: 12.2
Nvidia Driver Version: 535.104.12
Docker version 24.0.6, build ed223bc
ollama version 0.1.3

Any help is appreciated!

Originally created by @domWinter on GitHub (Oct 27, 2023). Original GitHub issue: https://github.com/ollama/ollama/issues/932 Originally assigned to: @BruceMacD on GitHub. Hi, I am trying to run ollama in a docker container with the nvidia runtime on a system with a Nvidia A100 and MIG enabled. When starting the ollama server I get the following error: `2023/10/27 09:16:05 routes.go:682: Warning: GPU support may not enabled, check you have installed install GPU drivers: failed to parse available VRAM: strconv.ParseInt: parsing "[Insufficient Permissions]": invalid syntax` The ollama server is running as root user. However, when running nvidia-smi inside the container I see the passed through GPU and MIG device and I am also able to allocate the GPU with e.g. pytorch in the same container. Cuda Version: 12.2 Nvidia Driver Version: 535.104.12 Docker version 24.0.6, build ed223bc ollama version 0.1.3 Any help is appreciated!
GiteaMirror commented 2026-04-28 02:19:15 -05:00
Author
Owner
Copy Link

@domWinter commented on GitHub (Oct 27, 2023):

Seems like the problematic call is e5d1ce4dde/llm/llama.go (L203)

nvidia-smi --query-gpu=memory.free --format=csv,noheader,nounits

Which returns [Insufficient Permissions]

<!-- gh-comment-id:1782615807 --> @domWinter commented on GitHub (Oct 27, 2023): Seems like the problematic call is https://github.com/jmorganca/ollama/blob/e5d1ce4dde0e30594c1b49d761ad98b50e3d17ab/llm/llama.go#L203 nvidia-smi --query-gpu=memory.free --format=csv,noheader,nounits Which returns [Insufficient Permissions]
GiteaMirror commented 2026-04-28 02:19:16 -05:00
Author
Owner
Copy Link

@BruceMacD commented on GitHub (Oct 27, 2023):

Thanks for opening the issue. Looks like the container might need some host level permissions it does not currently have. When you run ollama as the root user in the container it will have sudo permissions in the container, but not externally.

Another workaround is to manually set the num_gpu in a modelfile.

<!-- gh-comment-id:1783198547 --> @BruceMacD commented on GitHub (Oct 27, 2023): Thanks for opening the issue. Looks like the container might need some host level permissions it does not currently have. When you run ollama as the root user in the container it will have sudo permissions in the container, but not externally. Another workaround is to manually set the `num_gpu` in a modelfile.
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
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#46969
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?