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 #3112] Windows Error:pull model manifest return wsarecv: An existing connection was forcibly closed by the remote host. #27672

New Issue
Closed
opened 2026-04-22 05:11:43 -05:00 by GiteaMirror · 11 comments
GiteaMirror commented 2026-04-22 05:11:43 -05:00
Owner
Copy Link

Originally created by @heimu-liu on GitHub (Mar 13, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/3112

i can't down the model:
app.log
server.log

PS C:\Users\heimu\AppData\Local\Ollama> ollama pull llama2 pulling manifest Error: pull model manifest: Get "https://ollama.com/token?nonce=A-QmGZFS0za-Kv0GKrDy3Q&scope=repository%!A(MISSING)library%!F(MISSING)llama2%!A(MISSING)pull&service=ollama.com&ts=1710339130": read tcp 192.168.247.214:56798->34.120.132.20:443: wsarecv: An existing connection was forcibly closed by the remote host. PS C:\Users\heimu\AppData\Local\Ollama>

Originally created by @heimu-liu on GitHub (Mar 13, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/3112 i can't down the model: [app.log](https://github.com/ollama/ollama/files/14589977/app.log) [server.log](https://github.com/ollama/ollama/files/14589978/server.log) `PS C:\Users\heimu\AppData\Local\Ollama> ollama pull llama2 pulling manifest Error: pull model manifest: Get "https://ollama.com/token?nonce=A-QmGZFS0za-Kv0GKrDy3Q&scope=repository%!A(MISSING)library%!F(MISSING)llama2%!A(MISSING)pull&service=ollama.com&ts=1710339130": read tcp 192.168.247.214:56798->34.120.132.20:443: wsarecv: An existing connection was forcibly closed by the remote host. PS C:\Users\heimu\AppData\Local\Ollama>`
GiteaMirror added the networkingwindows labels 2026-04-22 05:11:43 -05:00
GiteaMirror commented 2026-04-22 05:11:46 -05:00
Author
Owner
Copy Link

@BruceMacD commented on GitHub (Mar 15, 2024):

Hi @heimu-liu this seems similar to #1149, are you using ollama behind a proxy? If so checkout the proxy configuration guide:
https://github.com/ollama/ollama/blob/main/docs/faq.md#how-do-i-use-ollama-behind-a-proxy

Hopefully that helps, let me know if that is the case.

<!-- gh-comment-id:1999449339 --> @BruceMacD commented on GitHub (Mar 15, 2024): Hi @heimu-liu this seems similar to #1149, are you using ollama behind a proxy? If so checkout the proxy configuration guide: https://github.com/ollama/ollama/blob/main/docs/faq.md#how-do-i-use-ollama-behind-a-proxy Hopefully that helps, let me know if that is the case.
GiteaMirror commented 2026-04-22 05:11:47 -05:00
Author
Owner
Copy Link

@heimu-liu commented on GitHub (Mar 16, 2024):

Contributor

Thank you very much for your reply!
I've tested it with both wireless and LTE, as well as proxy and non-proxy.
still get the same error.

<!-- gh-comment-id:2001448954 --> @heimu-liu commented on GitHub (Mar 16, 2024): > Contributor Thank you very much for your reply! I've tested it with both wireless and LTE, as well as proxy and non-proxy. still get the same error.
GiteaMirror commented 2026-04-22 05:11:48 -05:00
Author
Owner
Copy Link

@dhiltgen commented on GitHub (Mar 21, 2024):

Similar to #3191 and #2448

<!-- gh-comment-id:2012002487 --> @dhiltgen commented on GitHub (Mar 21, 2024): Similar to #3191 and #2448
GiteaMirror commented 2026-04-22 05:11:48 -05:00
Author
Owner
Copy Link

@Felixchen1024 commented on GitHub (Mar 28, 2024):

How can I fix this?

<!-- gh-comment-id:2024781325 --> @Felixchen1024 commented on GitHub (Mar 28, 2024): How can I fix this?
GiteaMirror commented 2026-04-22 05:11:49 -05:00
Author
Owner
Copy Link

@mxyng commented on GitHub (Mar 28, 2024):

Looking at the request both from the error in your description and the server access logs, it appears the request is getting mangled somehow.

This is being sent

https://ollama.com/token?nonce=A-QmGZFS0za-Kv0GKrDy3Q&scope=repository%!A(MISSING)library%!F(MISSING)llama2%!A(MISSING)pull&service=ollama.com&ts=1710339130

instead of this

https://ollama.com/token?nonce=A-QmGZFS0za-Kv0GKrDy3Q&scope=repository%3Alibrary%2Fllama2%3Apull&service=ollama.com&ts=1710339130

This should be caused by passing the real URL into fmt.Printf or similar functions: example.

Is it possible there's a man-in-the-middle proxy mangling the request?

<!-- gh-comment-id:2026153066 --> @mxyng commented on GitHub (Mar 28, 2024): Looking at the request both from the error in your description and the server access logs, it appears the request is getting mangled somehow. This is being sent ``` https://ollama.com/token?nonce=A-QmGZFS0za-Kv0GKrDy3Q&scope=repository%!A(MISSING)library%!F(MISSING)llama2%!A(MISSING)pull&service=ollama.com&ts=1710339130 ``` instead of this ``` https://ollama.com/token?nonce=A-QmGZFS0za-Kv0GKrDy3Q&scope=repository%3Alibrary%2Fllama2%3Apull&service=ollama.com&ts=1710339130 ``` This should be caused by passing the real URL into `fmt.Printf` or similar functions: [example](https://go.dev/play/p/V0sMmL70jIs). Is it possible there's a man-in-the-middle proxy mangling the request?
GiteaMirror commented 2026-04-22 05:11:50 -05:00
Author
Owner
Copy Link

@chopeen commented on GitHub (Mar 28, 2024):

My experiments in #2448 suggest that a correct URL is called and it is only mangled later, when logged.

I am facing the same problem when connected to a corporate network (Ubuntu 22.04, Ollama installed with Homebrew):

$ ollama run llava

pulling manifest 
Error: pull model manifest: Get "https://ollama.com/token?nonce=SW96RgmctcQmHJ37NXJ8KQ&scope=repository%!A(MISSING)library%!F(MISSING)llava%!A(MISSING)pull&service=ollama.com&ts=1711106784": read tcp 10.144.68.189:40600->34.120.132.20:443: read: connection reset by peer

The problem first appeared in v0.1.28.

When I downgrade to v0.1.27, I am able to pull the models again.

<!-- gh-comment-id:2026249945 --> @chopeen commented on GitHub (Mar 28, 2024): My experiments in #2448 suggest that a correct URL is called and it is only mangled later, when logged. --- I am facing the same problem **when connected to a corporate network** (Ubuntu 22.04, Ollama installed with Homebrew): ``` $ ollama run llava pulling manifest Error: pull model manifest: Get "https://ollama.com/token?nonce=SW96RgmctcQmHJ37NXJ8KQ&scope=repository%!A(MISSING)library%!F(MISSING)llava%!A(MISSING)pull&service=ollama.com&ts=1711106784": read tcp 10.144.68.189:40600->34.120.132.20:443: read: connection reset by peer ``` The problem first appeared in `v0.1.28`. When I downgrade to `v0.1.27`, I am able to pull the models again.
GiteaMirror commented 2026-04-22 05:11:51 -05:00
Author
Owner
Copy Link

@heimu-liu commented on GitHub (Mar 29, 2024):

My experiments in #2448 suggest that a correct URL is called and it is only mangled later, when logged.

I am facing the same problem when connected to a corporate network (Ubuntu 22.04, Ollama installed with Homebrew):

$ ollama run llava

pulling manifest 
Error: pull model manifest: Get "https://ollama.com/token?nonce=SW96RgmctcQmHJ37NXJ8KQ&scope=repository%!A(MISSING)library%!F(MISSING)llava%!A(MISSING)pull&service=ollama.com&ts=1711106784": read tcp 10.144.68.189:40600->34.120.132.20:443: read: connection reset by peer

The problem first appeared in v0.1.28.

When I downgrade to v0.1.27, I am able to pull the models again.

Thank you Very Much!

<!-- gh-comment-id:2026564134 --> @heimu-liu commented on GitHub (Mar 29, 2024): > My experiments in #2448 suggest that a correct URL is called and it is only mangled later, when logged. > > I am facing the same problem **when connected to a corporate network** (Ubuntu 22.04, Ollama installed with Homebrew): > > ``` > $ ollama run llava > > pulling manifest > Error: pull model manifest: Get "https://ollama.com/token?nonce=SW96RgmctcQmHJ37NXJ8KQ&scope=repository%!A(MISSING)library%!F(MISSING)llava%!A(MISSING)pull&service=ollama.com&ts=1711106784": read tcp 10.144.68.189:40600->34.120.132.20:443: read: connection reset by peer > ``` > > The problem first appeared in `v0.1.28`. > > When I downgrade to `v0.1.27`, I am able to pull the models again. Thank you Very Much!
GiteaMirror commented 2026-04-22 05:11:51 -05:00
Author
Owner
Copy Link

@heimu-liu commented on GitHub (Mar 29, 2024):

Problem solved by downgrade to v0.1.27

<!-- gh-comment-id:2026586878 --> @heimu-liu commented on GitHub (Mar 29, 2024): Problem solved by downgrade to v0.1.27
GiteaMirror commented 2026-04-22 05:11:52 -05:00
Author
Owner
Copy Link

@chopeen commented on GitHub (Mar 29, 2024):

@heimu-liu Downgrading is merely a workaround, but hopefully the problem can actually be solved in future version of Ollama.

@mxyng Do you prefer to reopen this issue or should I create a new one with all the information known so far?

<!-- gh-comment-id:2026844447 --> @chopeen commented on GitHub (Mar 29, 2024): @heimu-liu Downgrading is merely a workaround, but hopefully the problem can actually be solved in future version of Ollama. @mxyng Do you prefer to reopen this issue **or** should I create a new one with all the information known so far?
GiteaMirror commented 2026-04-22 05:11:53 -05:00
Author
Owner
Copy Link

@chopeen commented on GitHub (Apr 2, 2024):

I opened new issue #3452.

<!-- gh-comment-id:2031592911 --> @chopeen commented on GitHub (Apr 2, 2024): I opened new issue #3452.
GiteaMirror commented 2026-04-22 05:11:54 -05:00
Author
Owner
Copy Link

@nitinkr0411 commented on GitHub (Apr 23, 2024):

I'm facing the same issue with the latest version of ollama. Does not work in corporate proxy anymore.
Downgrading to v0.1.27 makes it work again.

<!-- gh-comment-id:2071437369 --> @nitinkr0411 commented on GitHub (Apr 23, 2024): I'm facing the same issue with the latest version of ollama. Does not work in corporate proxy anymore. Downgrading to v0.1.27 makes it work 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 networking windows
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#27672
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?