github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #7512] Snap 0.3.13 missing libcudart.so.12 #4779

New Issue
Closed
opened 2026-04-12 15:43:32 -05:00 by GiteaMirror · 5 comments
GiteaMirror commented 2026-04-12 15:43:32 -05:00
Owner
Copy Link

Originally created by @edmcman on GitHub (Nov 5, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/7512

What is the issue?

I am trying ollama for the first time. Since there was a snap available, I tried that first. I can download llama 3.2, but when I attempt to run it, I get:

(venv) ed@banana ~/P/re-copilot (dev)> ollama run llama3.2
Error: llama runner process has terminated: exit status 127

Here are the logs from snap logs ollama:

2024-11-05T10:38:44-05:00 ollama.listener[631587]: time=2024-11-05T10:38:44.915-05:00 level=INFO source=server.go:399 msg="starting llama server" cmd="/tmp/ollama2989640063/runners/cuda_v12/ollama_llama_server --model /var/snap/ollama/common/models/blobs/sha256-dde5aa3fc5ffc17176b5e8bdc82f587b24b2678c6c66101bf7da77af9f7ccdff --ctx-size 8192 --batch-size 512 --embedding --log-disable --n-gpu-layers 29 --parallel 4 --port 34001"
2024-11-05T10:38:44-05:00 ollama.listener[631587]: time=2024-11-05T10:38:44.915-05:00 level=INFO source=sched.go:449 msg="loaded runners" count=1
2024-11-05T10:38:44-05:00 ollama.listener[631587]: time=2024-11-05T10:38:44.915-05:00 level=INFO source=server.go:598 msg="waiting for llama runner to start responding"
2024-11-05T10:38:44-05:00 ollama.listener[631587]: time=2024-11-05T10:38:44.916-05:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server error"
2024-11-05T10:38:44-05:00 ollama.listener[631587]: /tmp/ollama2989640063/runners/cuda_v12/ollama_llama_server: error while loading shared libraries: libcudart.so.12: cannot open shared object file: No such file or directory
2024-11-05T10:38:45-05:00 ollama.listener[631587]: time=2024-11-05T10:38:45.166-05:00 level=ERROR source=sched.go:455 msg="error loading llama server" error="llama runner process has terminated: exit status 127"
2024-11-05T10:38:45-05:00 ollama.listener[631587]: [GIN] 2024/11/05 - 10:38:45 | 500 |  400.927983ms |       127.0.0.1 | POST     "/api/generate"
2024-11-05T10:38:50-05:00 ollama.listener[631587]: time=2024-11-05T10:38:50.312-05:00 level=WARN source=sched.go:646 msg="gpu VRAM usage didn't recover within timeout" seconds=5.145569241 model=/var/snap/ollama/common/models/blobs/sha256-dde5aa3fc5ffc17176b5e8bdc82f587b24b2678c6c66101bf7da77af9f7ccdff
2024-11-05T10:38:50-05:00 ollama.listener[631587]: time=2024-11-05T10:38:50.561-05:00 level=WARN source=sched.go:646 msg="gpu VRAM usage didn't recover within timeout" seconds=5.39517205 model=/var/snap/ollama/common/models/blobs/sha256-dde5aa3fc5ffc17176b5e8bdc82f587b24b2678c6c66101bf7da77af9f7ccdff
2024-11-05T10:38:50-05:00 ollama.listener[631587]: time=2024-11-05T10:38:50.811-05:00 level=WARN source=sched.go:646 msg="gpu VRAM usage didn't recover within timeout" seconds=5.644657486 model=/var/snap/ollama/common/models/blobs/sha256-dde5aa3fc5ffc17176b5e8bdc82f587b24b2678c6c66101bf7da77af9f7ccdff

It seems like the problem is that libcudart.so.12 is missing. Should it be getting that from the host, or the snap?

OS

Linux

GPU

Nvidia

CPU

Intel

Ollama version

0.3.13 snap

Originally created by @edmcman on GitHub (Nov 5, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/7512 ### What is the issue? I am trying ollama for the first time. Since there was a snap available, I tried that first. I can download llama 3.2, but when I attempt to run it, I get: ``` (venv) ed@banana ~/P/re-copilot (dev)> ollama run llama3.2 Error: llama runner process has terminated: exit status 127 ``` Here are the logs from `snap logs ollama`: ``` 2024-11-05T10:38:44-05:00 ollama.listener[631587]: time=2024-11-05T10:38:44.915-05:00 level=INFO source=server.go:399 msg="starting llama server" cmd="/tmp/ollama2989640063/runners/cuda_v12/ollama_llama_server --model /var/snap/ollama/common/models/blobs/sha256-dde5aa3fc5ffc17176b5e8bdc82f587b24b2678c6c66101bf7da77af9f7ccdff --ctx-size 8192 --batch-size 512 --embedding --log-disable --n-gpu-layers 29 --parallel 4 --port 34001" 2024-11-05T10:38:44-05:00 ollama.listener[631587]: time=2024-11-05T10:38:44.915-05:00 level=INFO source=sched.go:449 msg="loaded runners" count=1 2024-11-05T10:38:44-05:00 ollama.listener[631587]: time=2024-11-05T10:38:44.915-05:00 level=INFO source=server.go:598 msg="waiting for llama runner to start responding" 2024-11-05T10:38:44-05:00 ollama.listener[631587]: time=2024-11-05T10:38:44.916-05:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server error" 2024-11-05T10:38:44-05:00 ollama.listener[631587]: /tmp/ollama2989640063/runners/cuda_v12/ollama_llama_server: error while loading shared libraries: libcudart.so.12: cannot open shared object file: No such file or directory 2024-11-05T10:38:45-05:00 ollama.listener[631587]: time=2024-11-05T10:38:45.166-05:00 level=ERROR source=sched.go:455 msg="error loading llama server" error="llama runner process has terminated: exit status 127" 2024-11-05T10:38:45-05:00 ollama.listener[631587]: [GIN] 2024/11/05 - 10:38:45 | 500 | 400.927983ms | 127.0.0.1 | POST "/api/generate" 2024-11-05T10:38:50-05:00 ollama.listener[631587]: time=2024-11-05T10:38:50.312-05:00 level=WARN source=sched.go:646 msg="gpu VRAM usage didn't recover within timeout" seconds=5.145569241 model=/var/snap/ollama/common/models/blobs/sha256-dde5aa3fc5ffc17176b5e8bdc82f587b24b2678c6c66101bf7da77af9f7ccdff 2024-11-05T10:38:50-05:00 ollama.listener[631587]: time=2024-11-05T10:38:50.561-05:00 level=WARN source=sched.go:646 msg="gpu VRAM usage didn't recover within timeout" seconds=5.39517205 model=/var/snap/ollama/common/models/blobs/sha256-dde5aa3fc5ffc17176b5e8bdc82f587b24b2678c6c66101bf7da77af9f7ccdff 2024-11-05T10:38:50-05:00 ollama.listener[631587]: time=2024-11-05T10:38:50.811-05:00 level=WARN source=sched.go:646 msg="gpu VRAM usage didn't recover within timeout" seconds=5.644657486 model=/var/snap/ollama/common/models/blobs/sha256-dde5aa3fc5ffc17176b5e8bdc82f587b24b2678c6c66101bf7da77af9f7ccdff ``` It seems like the problem is that libcudart.so.12 is missing. Should it be getting that from the host, or the snap? ### OS Linux ### GPU Nvidia ### CPU Intel ### Ollama version 0.3.13 snap
GiteaMirror added the installlinuxbug labels 2026-04-12 15:43:32 -05:00
GiteaMirror commented 2026-04-12 15:43:34 -05:00
Author
Owner
Copy Link

@edmcman commented on GitHub (Nov 5, 2024):

I do seem to have cudart12.so on my host:

ed@banana ~/P/re-copilot (dev)> sudo ldconfig -p | fgrep cudart
        libcudart.so.12 (libc6,x86-64) => /usr/local/cuda/targets/x86_64-linux/lib/libcudart.so.12
        libcudart.so (libc6,x86-64) => /usr/local/cuda/targets/x86_64-linux/lib/libcudart.so
<!-- gh-comment-id:2457550791 --> @edmcman commented on GitHub (Nov 5, 2024): I do seem to have cudart12.so on my host: ``` ed@banana ~/P/re-copilot (dev)> sudo ldconfig -p | fgrep cudart libcudart.so.12 (libc6,x86-64) => /usr/local/cuda/targets/x86_64-linux/lib/libcudart.so.12 libcudart.so (libc6,x86-64) => /usr/local/cuda/targets/x86_64-linux/lib/libcudart.so ```
GiteaMirror commented 2026-04-12 15:43:34 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 5, 2024):

It's normally bundled with the application to prevent library version skew. snap is not the usual way to install ollama on linux, you may have more success with the canonical way: curl -fsSL https://ollama.com/install.sh | sh.

<!-- gh-comment-id:2457559562 --> @rick-github commented on GitHub (Nov 5, 2024): It's normally bundled with the application to prevent library version skew. snap is not the usual way to install ollama on linux, you may have more success with the canonical way: `curl -fsSL https://ollama.com/install.sh | sh`.
GiteaMirror commented 2026-04-12 15:43:35 -05:00
Author
Owner
Copy Link

@dhiltgen commented on GitHub (Nov 5, 2024):

The snap packaging isn't authored or maintained by the Ollama maintainers so unfortunately we wont be able to fix this. Hopefully the package identifies who created it and you can file an issue with them to update their packaging. Until then, as Rick suggested, you can uninstall the snap package and use our official curl install script.

<!-- gh-comment-id:2457670884 --> @dhiltgen commented on GitHub (Nov 5, 2024): The snap packaging isn't authored or maintained by the Ollama maintainers so unfortunately we wont be able to fix this. Hopefully the package identifies who created it and you can file an issue with them to update their packaging. Until then, as Rick suggested, you can uninstall the snap package and use our official curl install script.
GiteaMirror commented 2026-04-12 15:43:35 -05:00
Author
Owner
Copy Link

@edmcman commented on GitHub (Nov 5, 2024):

Thanks for clarifying. The snap's page linked to the ollama's issue page, so I thought that meant it was an official project.

<!-- gh-comment-id:2457708585 --> @edmcman commented on GitHub (Nov 5, 2024): Thanks for clarifying. The snap's page linked to the ollama's issue page, so I thought that meant it was an official project.
GiteaMirror commented 2026-04-12 15:43:36 -05:00
Author
Owner
Copy Link

@mz2 commented on GitHub (Nov 5, 2024):

Sorry about that, I'll correct the issue link to point at https://github.com/mz2/ollama-snap …

<!-- gh-comment-id:2457862389 --> @mz2 commented on GitHub (Nov 5, 2024): Sorry about that, I'll correct the issue link to point at https://github.com/mz2/ollama-snap …
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 install linux
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#4779
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?