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 #6098] Why is the llama3 model missing after I restart Ollama? When I run “ollama run llama3”, it re-pulls the manifest. #3814

New Issue
Closed
opened 2026-04-12 14:39:00 -05:00 by GiteaMirror · 6 comments
GiteaMirror commented 2026-04-12 14:39:00 -05:00
Owner
Copy Link

Originally created by @fanjikang on GitHub (Jul 31, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/6098

What is the issue?

Why is the llama3 model missing after I restart Ollama? When I run “ollama run llama3”, it re-pulls the manifest.

OS

Linux

GPU

Nvidia

CPU

Intel

Ollama version

v0.2.8

Originally created by @fanjikang on GitHub (Jul 31, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/6098 ### What is the issue? Why is the llama3 model missing after I restart Ollama? When I run “ollama run llama3”, it re-pulls the manifest. ### OS Linux ### GPU Nvidia ### CPU Intel ### Ollama version v0.2.8
GiteaMirror added the bug label 2026-04-12 14:39:00 -05:00
GiteaMirror commented 2026-04-12 14:39:01 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jul 31, 2024):

Did you change the server environment variable OLLAMA_MODELS? If you add server logs it may be easier to diagnose the issue.

<!-- gh-comment-id:2260733750 --> @rick-github commented on GitHub (Jul 31, 2024): Did you change the server environment variable `OLLAMA_MODELS`? If you add server logs it may be easier to diagnose the issue.
GiteaMirror commented 2026-04-12 14:39:01 -05:00
Author
Owner
Copy Link

@Siddharth-Latthe-07 commented on GitHub (Aug 2, 2024):

The above issue suggests that the model is not being persistently stored or there is a configuration problem causing it to be removed upon restart.
steps to diagnose:-

  1. Check Ollama Configuration
  2. Disk Space and Permissions
  3. Model cache and log directory
  4. service restarts and version updation
  5. Persistent Storage Configuration:
    ollama config set model_storage_path /path/to/persistent/storage
  6. Checking and Setting Up Model Cache Directory

Hope this helps,
Thanks

<!-- gh-comment-id:2265452851 --> @Siddharth-Latthe-07 commented on GitHub (Aug 2, 2024): The above issue suggests that the model is not being persistently stored or there is a configuration problem causing it to be removed upon restart. steps to diagnose:- 1. Check Ollama Configuration 2. Disk Space and Permissions 3. Model cache and log directory 4. service restarts and version updation 5. Persistent Storage Configuration: `ollama config set model_storage_path /path/to/persistent/storage` 6. Checking and Setting Up Model Cache Directory Hope this helps, Thanks
GiteaMirror commented 2026-04-12 14:39:01 -05:00
Author
Owner
Copy Link

@FellowTraveler commented on GitHub (Aug 12, 2024):

@fanjikang Can you please confirm via above tips if this is a real issue? Otherwise please close this issue.

<!-- gh-comment-id:2282958949 --> @FellowTraveler commented on GitHub (Aug 12, 2024): @fanjikang Can you please confirm via above tips if this is a real issue? Otherwise please close this issue.
GiteaMirror commented 2026-04-12 14:39:02 -05:00
Author
Owner
Copy Link

@skjortan23 commented on GitHub (Feb 6, 2025):

I am actually having the same problem. after a power outage ollama list is empty however i still see all my models by listing the model directory.

e.g

126G	./blobs
8,0K	./manifests/registry.ollama.ai/hengwen/watt-tool-8B
12K	./manifests/registry.ollama.ai/hengwen
8,0K	./manifests/registry.ollama.ai/library/llama3-groq-tool-use
8,0K	./manifests/registry.ollama.ai/library/codeqwen
8,0K	./manifests/registry.ollama.ai/library/phi4
8,0K	./manifests/registry.ollama.ai/library/Qwen2.5
12K	./manifests/registry.ollama.ai/library/qwen2.5-coder
8,0K	./manifests/registry.ollama.ai/library/deepseek-coder-v2
8,0K	./manifests/registry.ollama.ai/library/deepseek-coder
8,0K	./manifests/registry.ollama.ai/library/llama3.1
16K	./manifests/registry.ollama.ai/library/deepseek-r1
8,0K	./manifests/registry.ollama.ai/library/llama3.2
8,0K	./manifests/registry.ollama.ai/library/gemma2
8,0K	./manifests/registry.ollama.ai/library/yi-coder
8,0K	./manifests/registry.ollama.ai/library/phi3
8,0K	./manifests/registry.ollama.ai/library/mixtral
8,0K	./manifests/registry.ollama.ai/library/mistral
136K	./manifests/registry.ollama.ai/library
152K	./manifests/registry.ollama.ai
156K	./manifests
126G	./```
<!-- gh-comment-id:2639596461 --> @skjortan23 commented on GitHub (Feb 6, 2025): I am actually having the same problem. after a power outage `ollama list` is empty however i still see all my models by listing the model directory. e.g ```/usr/share/ollama/.ollama/models$ du -h ./ 126G ./blobs 8,0K ./manifests/registry.ollama.ai/hengwen/watt-tool-8B 12K ./manifests/registry.ollama.ai/hengwen 8,0K ./manifests/registry.ollama.ai/library/llama3-groq-tool-use 8,0K ./manifests/registry.ollama.ai/library/codeqwen 8,0K ./manifests/registry.ollama.ai/library/phi4 8,0K ./manifests/registry.ollama.ai/library/Qwen2.5 12K ./manifests/registry.ollama.ai/library/qwen2.5-coder 8,0K ./manifests/registry.ollama.ai/library/deepseek-coder-v2 8,0K ./manifests/registry.ollama.ai/library/deepseek-coder 8,0K ./manifests/registry.ollama.ai/library/llama3.1 16K ./manifests/registry.ollama.ai/library/deepseek-r1 8,0K ./manifests/registry.ollama.ai/library/llama3.2 8,0K ./manifests/registry.ollama.ai/library/gemma2 8,0K ./manifests/registry.ollama.ai/library/yi-coder 8,0K ./manifests/registry.ollama.ai/library/phi3 8,0K ./manifests/registry.ollama.ai/library/mixtral 8,0K ./manifests/registry.ollama.ai/library/mistral 136K ./manifests/registry.ollama.ai/library 152K ./manifests/registry.ollama.ai 156K ./manifests 126G ./```
GiteaMirror commented 2026-04-12 14:39:02 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Feb 6, 2025):

Did you change the server environment variable OLLAMA_MODELS? If you add server logs it may be easier to diagnose the issue.

<!-- gh-comment-id:2639600601 --> @rick-github commented on GitHub (Feb 6, 2025): Did you change the server environment variable `OLLAMA_MODELS`? If you add [server logs](https://github.com/ollama/ollama/blob/main/docs/troubleshooting.md#how-to-troubleshoot-issues) it may be easier to diagnose the issue.
GiteaMirror commented 2026-04-12 14:39:02 -05:00
Author
Owner
Copy Link

@FellowTraveler commented on GitHub (Feb 7, 2025):

You're probably aware of this but it's worth mentioning...

When I do "ollama list" I see this in the list:
llama3.3:70b-instruct-q5_K_M

...but q5_K_M is NOT the default quant. So if I do "ollama run llama3.3:70b-instruct-q5_K_M" it will run immediately.

But if I do "ollama run llama3.3" it will start downloading q4_0 or whatever the default quant is.

<!-- gh-comment-id:2642478663 --> @FellowTraveler commented on GitHub (Feb 7, 2025): You're probably aware of this but it's worth mentioning... When I do "ollama list" I see this in the list: llama3.3:70b-instruct-q5_K_M ...but q5_K_M is NOT the default quant. So if I do "ollama run llama3.3:70b-instruct-q5_K_M" it will run immediately. **But if I do "ollama run llama3.3" it will start downloading q4_0 or whatever the default quant is.**
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#3814
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?