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

[GH-ISSUE #14658] ollama docker loses access to nvidia gpu #35252

New Issue
Closed
opened 2026-04-22 19:38:24 -05:00 by GiteaMirror · 3 comments
GiteaMirror commented 2026-04-22 19:38:24 -05:00
Owner
Copy Link

Originally created by @ACodingfreak on GitHub (Mar 6, 2026).
Original GitHub issue: https://github.com/ollama/ollama/issues/14658

What is the issue?

As shown in below logs, after sometime ollama running as a docker container is not able to access the underlying NVIDIA GPU as shown below. Eventhough I am able to access the same GPUs by running nvidia-smi from the host machine. If I stop and restart the same ollama container everything work fine as expected.

Relevant log output

$ sudo docker exec -it ollama nvidia-smi
Failed to initialize NVML: Unknown Error

$ nvidia-smi
Thu Mar  5 20:45:27 2026
+---------------------------------------------------------------------------------------+
| NVIDIA-SMI 535.288.01             Driver Version: 535.288.01   CUDA Version: 12.2     |
|-----------------------------------------+----------------------+----------------------+
| GPU  Name                 Persistence-M | Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |         Memory-Usage | GPU-Util  Compute M. |
|                                         |                      |               MIG M. |
|=========================================+======================+======================|
|   0  NVIDIA A30                     Off | 00000000:9B:00.0 Off |                    0 |
| N/A   33C    P0              28W / 165W |      0MiB / 24576MiB |      0%      Default |
|                                         |                      |             Disabled |
+-----------------------------------------+----------------------+----------------------+

+---------------------------------------------------------------------------------------+
| Processes:                                                                            |
|  GPU   GI   CI        PID   Type   Process name                            GPU Memory |
|        ID   ID                                                             Usage      |
|=======================================================================================|
|  No running processes found                                                           |
+---------------------------------------------------------------------------------------+

$ sudo docker exec -it ollama ollama --version
ollama version is 0.15.4

$ sudo docker stop ollama
ollama
$ sudo docker rm ollama
ollama
$sudo docker run -d --gpus=all -v ollama:/root/.ollama -p 11434:11434 --name ollama ollama/ollama
0815be239db8b142db97f6d67571db8f8af99156162777a141c76c226b5d04c7
$ sudo docker exec -it ollama nvidia-smi
Fri Mar  6 04:54:05 2026
+---------------------------------------------------------------------------------------+
| NVIDIA-SMI 535.288.01             Driver Version: 535.288.01   CUDA Version: 12.2     |
|-----------------------------------------+----------------------+----------------------+
| GPU  Name                 Persistence-M | Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |         Memory-Usage | GPU-Util  Compute M. |
|                                         |                      |               MIG M. |
|=========================================+======================+======================|
|   0  NVIDIA A30                     Off | 00000000:9B:00.0 Off |                    0 |
| N/A   33C    P0              28W / 165W |      0MiB / 24576MiB |      0%      Default |
|                                         |                      |             Disabled |
+-----------------------------------------+----------------------+----------------------+

+---------------------------------------------------------------------------------------+
| Processes:                                                                            |
|  GPU   GI   CI        PID   Type   Process name                            GPU Memory |
|        ID   ID                                                             Usage      |
|=======================================================================================|
|  No running processes found                                                           |
+---------------------------------------------------------------------------------------+

OS

Docker

GPU

Nvidia

CPU

No response

Ollama version

0.15.4

Originally created by @ACodingfreak on GitHub (Mar 6, 2026). Original GitHub issue: https://github.com/ollama/ollama/issues/14658 ### What is the issue? As shown in below logs, after sometime ollama running as a docker container is not able to access the underlying NVIDIA GPU as shown below. Eventhough I am able to access the same GPUs by running nvidia-smi from the host machine. If I stop and restart the same ollama container everything work fine as expected. ### Relevant log output ```shell $ sudo docker exec -it ollama nvidia-smi Failed to initialize NVML: Unknown Error $ nvidia-smi Thu Mar 5 20:45:27 2026 +---------------------------------------------------------------------------------------+ | NVIDIA-SMI 535.288.01 Driver Version: 535.288.01 CUDA Version: 12.2 | |-----------------------------------------+----------------------+----------------------+ | GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. | | | | MIG M. | |=========================================+======================+======================| | 0 NVIDIA A30 Off | 00000000:9B:00.0 Off | 0 | | N/A 33C P0 28W / 165W | 0MiB / 24576MiB | 0% Default | | | | Disabled | +-----------------------------------------+----------------------+----------------------+ +---------------------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=======================================================================================| | No running processes found | +---------------------------------------------------------------------------------------+ $ sudo docker exec -it ollama ollama --version ollama version is 0.15.4 $ sudo docker stop ollama ollama $ sudo docker rm ollama ollama $sudo docker run -d --gpus=all -v ollama:/root/.ollama -p 11434:11434 --name ollama ollama/ollama 0815be239db8b142db97f6d67571db8f8af99156162777a141c76c226b5d04c7 $ sudo docker exec -it ollama nvidia-smi Fri Mar 6 04:54:05 2026 +---------------------------------------------------------------------------------------+ | NVIDIA-SMI 535.288.01 Driver Version: 535.288.01 CUDA Version: 12.2 | |-----------------------------------------+----------------------+----------------------+ | GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. | | | | MIG M. | |=========================================+======================+======================| | 0 NVIDIA A30 Off | 00000000:9B:00.0 Off | 0 | | N/A 33C P0 28W / 165W | 0MiB / 24576MiB | 0% Default | | | | Disabled | +-----------------------------------------+----------------------+----------------------+ +---------------------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=======================================================================================| | No running processes found | +---------------------------------------------------------------------------------------+ ``` ### OS Docker ### GPU Nvidia ### CPU _No response_ ### Ollama version 0.15.4
GiteaMirror added the bug label 2026-04-22 19:38:24 -05:00
GiteaMirror commented 2026-04-22 19:38:25 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Mar 6, 2026):

https://github.com/ollama/ollama/blob/main/docs/troubleshooting.mdx#linux-docker

<!-- gh-comment-id:4010081152 --> @rick-github commented on GitHub (Mar 6, 2026): https://github.com/ollama/ollama/blob/main/docs/troubleshooting.mdx#linux-docker
GiteaMirror commented 2026-04-22 19:38:26 -05:00
Author
Owner
Copy Link

@ACodingfreak commented on GitHub (Mar 7, 2026):

Thanks for the reply @rick-github

As shown below I am using the latest docker version with cgroupv2 and driver from systemd. The change which you have suggested is to move driver to cgroupfs. I can give a try for the same but curious why it makes a difference ?

$ sudo docker info | grep -i -E 'server version|cgroup|runtimes'
[sudo] password for ipi:
 Server Version: 28.5.1
 Cgroup Driver: systemd
 Cgroup Version: 2
 Runtimes: io.containerd.runc.v2 runc
  cgroupns
<!-- gh-comment-id:4014869310 --> @ACodingfreak commented on GitHub (Mar 7, 2026): Thanks for the reply @rick-github As shown below I am using the latest docker version with cgroupv2 and driver from systemd. The change which you have suggested is to move driver to cgroupfs. I can give a try for the same but curious why it makes a difference ? ``` $ sudo docker info | grep -i -E 'server version|cgroup|runtimes' [sudo] password for ipi: Server Version: 28.5.1 Cgroup Driver: systemd Cgroup Version: 2 Runtimes: io.containerd.runc.v2 runc cgroupns ```
GiteaMirror commented 2026-04-22 19:38:26 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Mar 7, 2026):

https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/troubleshooting.html#mitigations-and-workarounds

<!-- gh-comment-id:4016219562 --> @rick-github commented on GitHub (Mar 7, 2026): https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/troubleshooting.html#mitigations-and-workarounds
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
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#35252
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?