github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-08 00:51:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #7522] not able to download models from ollama behind proxy #4783

New Issue
Closed
opened 2026-04-12 15:43:52 -05:00 by GiteaMirror · 19 comments
GiteaMirror commented 2026-04-12 15:43:52 -05:00
Owner
Copy Link

Originally created by @anshika1234 on GitHub (Nov 6, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/7522

What is the issue?

ollama run qwen:4b
pulling manifest
pulling 46bb65206e0e... 0% ▕ ▏ 3.3 MB/2.3 GB 1.4 KB/s 99h+^****

-------------------------------------------- Log content -------------------
ollama[719]: time=2024-11-06T10:08:16.817+05:30 level=INFO source=download.go:370 msg="8eeb52dfb3bb part 15 stalled; retrying. If this persists, press ctrl-c to exit, then 'ollama pull' to find a faster connection."

-----------------------------------------------Settings-----------
https_proxy setup is done in ollama environment

Please suggest how to debug further.

OS

Linux

GPU

No response

CPU

Intel

Ollama version

0.3.14

Originally created by @anshika1234 on GitHub (Nov 6, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/7522 ### What is the issue? ollama run qwen:4b pulling manifest pulling 46bb65206e0e... 0% ▕ ▏ 3.3 MB/2.3 GB 1.4 KB/s 99h+^**** -------------------------------------------- Log content ------------------- ollama[719]: time=2024-11-06T10:08:16.817+05:30 level=INFO source=download.go:370 msg="8eeb52dfb3bb part 15 stalled; retrying. If this persists, press ctrl-c to exit, then 'ollama pull' to find a faster connection." -----------------------------------------------Settings----------- https_proxy setup is done in ollama environment Please suggest how to debug further. ### OS Linux ### GPU _No response_ ### CPU Intel ### Ollama version 0.3.14
GiteaMirror added the bug label 2026-04-12 15:43:52 -05:00
GiteaMirror commented 2026-04-12 15:43:54 -05:00
Author
Owner
Copy Link

@nikhil-swamix commented on GitHub (Nov 6, 2024):

could you try a different proxy? some proxies disallow ports, or have bandwidth throttles in my exp... let me know if work

<!-- gh-comment-id:2458815800 --> @nikhil-swamix commented on GitHub (Nov 6, 2024): could you try a different proxy? some proxies disallow ports, or have bandwidth throttles in my exp... let me know if work
GiteaMirror commented 2026-04-12 15:43:55 -05:00
Author
Owner
Copy Link

@anshika1234 commented on GitHub (Nov 6, 2024):

No i cannot change the proxy. But I confirmed there is no port blocking in proxy server.

Need to debug it further. How do it?

<!-- gh-comment-id:2458875937 --> @anshika1234 commented on GitHub (Nov 6, 2024): No i cannot change the proxy. But I confirmed there is no port blocking in proxy server. Need to debug it further. How do it?
GiteaMirror commented 2026-04-12 15:43:55 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 6, 2024):

Server logs may help in debugging.

What's the output of the commands:

systemctl cat --no-pager ollama
curl -D - https://registry.ollama.ai/v2/library/gemma2/manifests/2b
curl -D - https://registry.ollama.ai/v2/library/gemma2/manifests/2b --noproxy "*"
<!-- gh-comment-id:2458898306 --> @rick-github commented on GitHub (Nov 6, 2024): [Server logs](https://github.com/ollama/ollama/blob/main/docs/troubleshooting.md#how-to-troubleshoot-issues) may help in debugging. What's the output of the commands: ``` systemctl cat --no-pager ollama curl -D - https://registry.ollama.ai/v2/library/gemma2/manifests/2b curl -D - https://registry.ollama.ai/v2/library/gemma2/manifests/2b --noproxy "*" ```
GiteaMirror commented 2026-04-12 15:43:56 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 6, 2024):

Server logs?

<!-- gh-comment-id:2458911917 --> @rick-github commented on GitHub (Nov 6, 2024): Server logs?
GiteaMirror commented 2026-04-12 15:43:57 -05:00
Author
Owner
Copy Link

@anshika1234 commented on GitHub (Nov 6, 2024):

curl -D - https://registry.ollama.ai/v2/library/gemma2/manifests/2b --noproxy "*"
curl: (28) Failed to connect to registry.ollama.ai port 443 after 217352 ms: Connection timed out

-------------------------------------------- Log content -------------------
level=INFO source=download.go:370 msg="8eeb52dfb3bb part 3 stalled; retrying. If this persists, press ctrl-c to exit, then 'ollama pull' to find a faster connection."

<!-- gh-comment-id:2458920864 --> @anshika1234 commented on GitHub (Nov 6, 2024): curl -D - https://registry.ollama.ai/v2/library/gemma2/manifests/2b --noproxy "*" curl: (28) Failed to connect to registry.ollama.ai port 443 after 217352 ms: Connection timed out -------------------------------------------- Log content ------------------- level=INFO source=download.go:370 msg="8eeb52dfb3bb part 3 stalled; retrying. If this persists, press ctrl-c to exit, then 'ollama pull' to find a faster connection."
GiteaMirror commented 2026-04-12 15:43:57 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 6, 2024):

Full server log.

<!-- gh-comment-id:2458922552 --> @rick-github commented on GitHub (Nov 6, 2024): Full server log.
GiteaMirror commented 2026-04-12 15:43:58 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 6, 2024):

Full log, there is information in other parts that may be useful in debugging the problem you have.

<!-- gh-comment-id:2458929793 --> @rick-github commented on GitHub (Nov 6, 2024): Full log, there is information in other parts that may be useful in debugging the problem you have.
GiteaMirror commented 2026-04-12 15:43:59 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 6, 2024):

Is the ollama server running on the same machine you are running the ollama run command on?

<!-- gh-comment-id:2458935940 --> @rick-github commented on GitHub (Nov 6, 2024): Is the ollama server running on the same machine you are running the `ollama run` command on?
GiteaMirror commented 2026-04-12 15:44:00 -05:00
Author
Owner
Copy Link

@anshika1234 commented on GitHub (Nov 6, 2024):

yes

<!-- gh-comment-id:2458936622 --> @anshika1234 commented on GitHub (Nov 6, 2024): yes
GiteaMirror commented 2026-04-12 15:44:01 -05:00
Author
Owner
Copy Link

@nikhil-swamix commented on GitHub (Nov 6, 2024):

try this just for sample...
if this work, i believe my guess would be right. confirm result once done.
ollama run qwen2.5:0.5b

<!-- gh-comment-id:2458951340 --> @nikhil-swamix commented on GitHub (Nov 6, 2024): try this just for sample... if this work, i believe my guess would be right. confirm result once done. `ollama run qwen2.5:0.5b`
GiteaMirror commented 2026-04-12 15:44:01 -05:00
Author
Owner
Copy Link

@anshika1234 commented on GitHub (Nov 6, 2024):

/usr/share/ollama/ollama2477802559/runners/cpu# ollama run qwen2.5:0.5b
pulling manifest
pulling c5396e06af29... 100% ▕██████████████████████████▏ 397 MB
pulling 66b9ea09bd5b... 100% ▕██████████████████████████▏ 68 B
pulling eb4402837c78... 100% ▕██████████████████████████▏ 1.5 KB
pulling 832dd9e00a68... 100% ▕██████████████████████████▏ 11 KB
pulling 005f95c74751... 100% ▕██████████████████████████▏ 490 B
verifying sha256 digest
Error: digest mismatch, file must be downloaded again: want sha256:c5396e06af294bd101b30dce59131a76d2b773e76950acc870eda801d3ab0515, got sha256:5bdde73520ef51f67ce75262ffe51b46aca27e89ba7647eb0f5c82cdd81fa479

what next?

<!-- gh-comment-id:2458958355 --> @anshika1234 commented on GitHub (Nov 6, 2024): /usr/share/ollama/ollama2477802559/runners/cpu# ollama run qwen2.5:0.5b pulling manifest pulling c5396e06af29... 100% ▕██████████████████████████▏ 397 MB pulling 66b9ea09bd5b... 100% ▕██████████████████████████▏ 68 B pulling eb4402837c78... 100% ▕██████████████████████████▏ 1.5 KB pulling 832dd9e00a68... 100% ▕██████████████████████████▏ 11 KB pulling 005f95c74751... 100% ▕██████████████████████████▏ 490 B verifying sha256 digest Error: digest mismatch, file must be downloaded again: want sha256:c5396e06af294bd101b30dce59131a76d2b773e76950acc870eda801d3ab0515, got sha256:5bdde73520ef51f67ce75262ffe51b46aca27e89ba7647eb0f5c82cdd81fa479 _______________________________________ what next?
GiteaMirror commented 2026-04-12 15:44:02 -05:00
Author
Owner
Copy Link

@nikhil-swamix commented on GitHub (Nov 6, 2024):

thers problem with your I/O, it could be disk space issue, faulty drive, or if in container, some settings/permissions.
previously i discovered that, it try to allocate full block area on disk, before it download. my m2 drive was having issue.
so best attempt would be to change the OLLAMA_MODELS directory and try the same run command.
refer variable settings #7523
do df -h
select a big drive as path. and set as OLLAMA_MODELS
OLLAMA_TMPDIR=same as OLLAMA models
and make sure write permissions are correct.
on a side note, try with only recent models, as the ollama registery may cold archive older model which you initially tried...

<!-- gh-comment-id:2459012383 --> @nikhil-swamix commented on GitHub (Nov 6, 2024): thers problem with your I/O, it could be disk space issue, faulty drive, or if in container, some settings/permissions. previously i discovered that, it try to allocate full block area on disk, before it download. my m2 drive was having issue. so best attempt would be to change the OLLAMA_MODELS directory and try the same run command. refer variable settings #7523 do `df -h` select a big drive as path. and set as OLLAMA_MODELS `OLLAMA_TMPDIR=same as OLLAMA models` and make sure write permissions are correct. on a side note, try with only recent models, as the ollama registery may cold archive older model which you initially tried...
GiteaMirror commented 2026-04-12 15:44:03 -05:00
Author
Owner
Copy Link

@nikhil-swamix commented on GitHub (Nov 6, 2024):

Attempt 001:

# Get and set user's .ollama directory explicitly
OLLAMA_HOME=$(eval echo ~$USER)/.ollama
# Create directory if needed
mkdir -p $OLLAMA_HOME
mkdir -p $OLLAMA_HOME/models
# Create ollama user and group if they don't exist
sudo groupadd -f ollama 
sudo useradd -r -g ollama ollama 2>/dev/null || true
# Add current user to ollama group
sudo usermod -aG ollama $USER
# Set correct ownership and permissions
sudo chown -R ollama:ollama $OLLAMA_HOME
sudo chmod -R 770 $OLLAMA_HOME
# Stop ollama service safely  
sudo service ollama stop
# Set model and temp directories
export OLLAMA_MODELS=$OLLAMA_HOME/models
export OLLAMA_TEMPDIR=$OLLAMA_HOME/models
# Start ollama with explicit home
OLLAMA_HOME=$OLLAMA_HOME OLLAMA_MODELS=$OLLAMA_MODELS OLLAMA_TEMPDIR=$OLLAMA_TEMPDIR ollama run qwen2.5:0.5b
# Note: You may need to log out and back in for group changes to take effect
echo "Please log out and log back in for group changes to take effect"

Notes: Previous ollama must die for new changes to effect, modify the last part as required.

<!-- gh-comment-id:2459305835 --> @nikhil-swamix commented on GitHub (Nov 6, 2024): Attempt 001: ```bash # Get and set user's .ollama directory explicitly OLLAMA_HOME=$(eval echo ~$USER)/.ollama # Create directory if needed mkdir -p $OLLAMA_HOME mkdir -p $OLLAMA_HOME/models # Create ollama user and group if they don't exist sudo groupadd -f ollama sudo useradd -r -g ollama ollama 2>/dev/null || true # Add current user to ollama group sudo usermod -aG ollama $USER # Set correct ownership and permissions sudo chown -R ollama:ollama $OLLAMA_HOME sudo chmod -R 770 $OLLAMA_HOME # Stop ollama service safely sudo service ollama stop # Set model and temp directories export OLLAMA_MODELS=$OLLAMA_HOME/models export OLLAMA_TEMPDIR=$OLLAMA_HOME/models # Start ollama with explicit home OLLAMA_HOME=$OLLAMA_HOME OLLAMA_MODELS=$OLLAMA_MODELS OLLAMA_TEMPDIR=$OLLAMA_TEMPDIR ollama run qwen2.5:0.5b # Note: You may need to log out and back in for group changes to take effect echo "Please log out and log back in for group changes to take effect" ``` Notes: Previous ollama must die for new changes to effect, modify the last part as required.
GiteaMirror commented 2026-04-12 15:44:03 -05:00
Author
Owner
Copy Link

@anshika1234 commented on GitHub (Nov 6, 2024):

bypassing proxy solved the problem. Can we conclude it is not I/O problem. Thanks a lot for the responses.

<!-- gh-comment-id:2459403574 --> @anshika1234 commented on GitHub (Nov 6, 2024): bypassing proxy solved the problem. Can we conclude it is not I/O problem. Thanks a lot for the responses.
GiteaMirror commented 2026-04-12 15:44:04 -05:00
Author
Owner
Copy Link

@nikhil-swamix commented on GitHub (Nov 6, 2024):

Dear Anshika, glad to hear that the attempts yielded positive outcome. hope issue marked as closed.
best regards,
Swamix Global AI Solutions, swamix.com

<!-- gh-comment-id:2459411046 --> @nikhil-swamix commented on GitHub (Nov 6, 2024): Dear Anshika, glad to hear that the attempts yielded positive outcome. hope issue marked as closed. best regards, Swamix Global AI Solutions, swamix.com
GiteaMirror commented 2026-04-12 15:44:05 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 6, 2024):

OLLAMA_HOME=$OLLAMA_HOME OLLAMA_MODELS=$OLLAMA_MODELS OLLAMA_TEMPDIR=$OLLAMA_TEMPDIR ollama run qwen2.5:0.5b

This will not change the environment variables, they need to be set in the server environment.

<!-- gh-comment-id:2459422538 --> @rick-github commented on GitHub (Nov 6, 2024): > `OLLAMA_HOME=$OLLAMA_HOME OLLAMA_MODELS=$OLLAMA_MODELS OLLAMA_TEMPDIR=$OLLAMA_TEMPDIR ollama run qwen2.5:0.5b` This will not change the environment variables, they need to be set in the [server environment](https://github.com/ollama/ollama/blob/main/docs/faq.md#how-do-i-configure-ollama-server).
GiteaMirror commented 2026-04-12 15:44:05 -05:00
Author
Owner
Copy Link

@nikhil-swamix commented on GitHub (Nov 6, 2024):

OLLAMA_HOME=$OLLAMA_HOME OLLAMA_MODELS=$OLLAMA_MODELS OLLAMA_TEMPDIR=$OLLAMA_TEMPDIR ollama run qwen2.5:0.5b

This will not change the environment variables, they need to be set in the server environment.

ahh, agreed, in windows powershell $env:something=value worked, my answer may have been biased from debugging standpoint. this one i believe it would set only one time per session and lost . nevertheless, thanks for correction.

<!-- gh-comment-id:2459433488 --> @nikhil-swamix commented on GitHub (Nov 6, 2024): > > `OLLAMA_HOME=$OLLAMA_HOME OLLAMA_MODELS=$OLLAMA_MODELS OLLAMA_TEMPDIR=$OLLAMA_TEMPDIR ollama run qwen2.5:0.5b` > > This will not change the environment variables, they need to be set in the [server environment](https://github.com/ollama/ollama/blob/main/docs/faq.md#how-do-i-configure-ollama-server). ahh, agreed, in windows powershell `$env:something=value` worked, my answer may have been biased from debugging standpoint. this one i believe it would set only one time per session and lost . nevertheless, thanks for correction.
GiteaMirror commented 2026-04-12 15:44:06 -05:00
Author
Owner
Copy Link

@anshika1234 commented on GitHub (Nov 6, 2024):

OLLAMA_HOME=$OLLAMA_HOME OLLAMA_MODELS=$OLLAMA_MODELS OLLAMA_TEMPDIR=$OLLAMA_TEMPDIR ollama run qwen2.5:0.5b

This will not change the environment variables, they need to be set in the server environment.

yes i have set in server enviroment only. But it worked only when proxy was bypassed.

<!-- gh-comment-id:2459455610 --> @anshika1234 commented on GitHub (Nov 6, 2024): > > `OLLAMA_HOME=$OLLAMA_HOME OLLAMA_MODELS=$OLLAMA_MODELS OLLAMA_TEMPDIR=$OLLAMA_TEMPDIR ollama run qwen2.5:0.5b` > > This will not change the environment variables, they need to be set in the [server environment](https://github.com/ollama/ollama/blob/main/docs/faq.md#how-do-i-configure-ollama-server). yes i have set in server enviroment only. But it worked only when proxy was bypassed.
GiteaMirror commented 2026-04-12 15:44:06 -05:00
Author
Owner
Copy Link

@anshika1234 commented on GitHub (Nov 6, 2024):

Thanks again.

<!-- gh-comment-id:2459456071 --> @anshika1234 commented on GitHub (Nov 6, 2024): Thanks again.
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#4783
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?