github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-24 06:31:50 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #10101] Update broken on Linux #84307

New Issue
Closed
opened 2026-05-09 20:25:25 -05:00 by GiteaMirror · 3 comments
GiteaMirror commented 2026-05-09 20:25:25 -05:00
Owner
Copy Link

Originally created by @Mugane on GitHub (Apr 3, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/10101

What is the issue?

Running curl -fsSL https://ollama.com/install.sh | sh does not upgrade properly. Previously running with GPU detected, no issues, now:

  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
  0     0    0     0    0     0      0      0 --:--:-- --:--:-- --:--:--     0>>> Installing ollama to /usr/local
100 13281    0 13281    0     0  16896      0 --:--:-- --:--:-- --:--:-- 16875
>>> Downloading Linux amd64 bundle
######################################################################## 100.0%
WARNING: Unable to detect NVIDIA/AMD GPU. Install lspci or lshw to automatically detect and install GPU dependencies.
>>> The Ollama API is now available at 127.0.0.1:11434.
>>> Install complete. Run "ollama" from the command line.

It has failed to detect the GPU. It has failed to restart. Maybe it has failed in other ways too. It has basically failed completely. If I can't upgrade I'll just rebuild the docker container, but I should be able to upgrade ollama in place (in the running docker instance). What if I don't have the capacity to take down the entire server while it happens? What if I don't trust the CI/CD guys to produce a new image for the entire system that is backwards-compatible? What if I just want to try the new version before committing to the change in my own CI/CD process? This is a pretty basic need.

Relevant log output

OS

Docker

GPU

Nvidia

CPU

Intel

Ollama version

0.6.2

Originally created by @Mugane on GitHub (Apr 3, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/10101 ### What is the issue? Running `curl -fsSL https://ollama.com/install.sh | sh` does not upgrade properly. Previously running with GPU detected, no issues, now: ``` % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0>>> Installing ollama to /usr/local 100 13281 0 13281 0 0 16896 0 --:--:-- --:--:-- --:--:-- 16875 >>> Downloading Linux amd64 bundle ######################################################################## 100.0% WARNING: Unable to detect NVIDIA/AMD GPU. Install lspci or lshw to automatically detect and install GPU dependencies. >>> The Ollama API is now available at 127.0.0.1:11434. >>> Install complete. Run "ollama" from the command line. ``` It has failed to detect the GPU. It has failed to restart. Maybe it has failed in other ways too. It has basically failed completely. If I can't upgrade I'll just rebuild the docker container, but I should be able to upgrade ollama in place (in the running docker instance). What if I don't have the capacity to take down the entire server while it happens? What if I don't trust the CI/CD guys to produce a new image for the entire system that is backwards-compatible? What if I just want to try the new version before committing to the change in my own CI/CD process? This is a pretty basic need. ### Relevant log output ```shell ``` ### OS Docker ### GPU Nvidia ### CPU Intel ### Ollama version 0.6.2
GiteaMirror added the bug label 2026-05-09 20:25:25 -05:00
GiteaMirror commented 2026-05-09 20:25:30 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Apr 3, 2025):

install.sh is for running on a system with systemd and tools for probing the environment. If you just want to swap in a new binary, use the ollama-linux-amd64.tgz tarball as described here. Not sure why you'd upgrade inside a docker container instead of just pulling (or inheriting from, if you have a custom Dockerfile) from the ollama docker image.

<!-- gh-comment-id:2775285600 --> @rick-github commented on GitHub (Apr 3, 2025): install.sh is for running on a system with systemd and tools for probing the environment. If you just want to swap in a new binary, use the ollama-linux-amd64.tgz tarball as described [here](https://github.com/ollama/ollama/blob/main/docs/linux.md#manual-install). Not sure why you'd upgrade inside a docker container instead of just pulling (or inheriting from, if you have a custom Dockerfile) from the ollama docker image.
GiteaMirror commented 2026-05-09 20:25:32 -05:00
Author
Owner
Copy Link

@Mugane commented on GitHub (Apr 7, 2025):

Re: Why upgrade in Docker: I have already explained this (see above). There are always good reasons that you haven't thought of. Mine will be different from someone else's. The reason should not be relevant - it cannot be upgraded as described or the documentation is incomplete, regardless of motivation.

Those linked instructions are for a manual install. Which of those steps are sufficient when I'm only upgrading? We're trying to avoid a reinstall (that's the whole point of upgrading instead of starting from scratch). Also the install.sh script shouldn't need to probe the system at all - if we're upgrading, we already have an installation that has knowledge of the system. Why repeat that? Or does it basically just do a fresh install over the existing system?

<!-- gh-comment-id:2781923073 --> @Mugane commented on GitHub (Apr 7, 2025): Re: Why upgrade in Docker: I have already explained this (see above). There are always good reasons that you haven't thought of. Mine will be different from someone else's. The reason should not be relevant - it cannot be upgraded as described or the documentation is incomplete, regardless of motivation. Those linked instructions are for a manual install. Which of those steps are sufficient when I'm only upgrading? We're trying to avoid a reinstall (that's the whole point of upgrading instead of starting from scratch). Also the install.sh script shouldn't need to probe the system at all - if we're upgrading, we already have an installation that has knowledge of the system. Why repeat that? Or does it basically just do a fresh install over the existing system?
GiteaMirror commented 2026-05-09 20:25:33 -05:00
Author
Owner
Copy Link

@Mugane commented on GitHub (Apr 13, 2025):

@ghmer could you please explain your motivation for the downvote?

<!-- gh-comment-id:2799958400 --> @Mugane commented on GitHub (Apr 13, 2025): @ghmer could you please explain your motivation for the downvote?
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#84307
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?