github-starred/ollama-ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-04-27 07:57:18 -05:00
Code Issues 2.2k Packages Projects Releases 100 Wiki Activity

Ollama on Windows occupied all available ports when downloading #4416

New Issue
Closed
opened 2025-11-12 12:18:59 -06:00 by GiteaMirror · 7 comments
GiteaMirror commented 2025-11-12 12:18:59 -06:00
Owner
Copy Link

Originally created by @TheStarAlight on GitHub (Sep 26, 2024).

What is the issue?

When I'm trying to download a model from Ollama for Windows, after a while, my browsers cannot visit any other website, showing "connection refused". And the download would also fail (after the first part of this model finished, the next part cannot start and reports error).
The log ~/AppData/Local/Ollama/server.log shows Error: listen tcp 127.0.0.1:11434: bind: Only one usage of each socket address (protocol/network address/port) is normally permitted.
I also used the netstat, which shows that port number up to 65535 is occupied by ollama.
Is there any way to solve it?

OS

Windows

GPU

Nvidia

CPU

AMD

Ollama version

Win 0.3.12

Originally created by @TheStarAlight on GitHub (Sep 26, 2024). ### What is the issue? When I'm trying to download a model from Ollama for Windows, after a while, my browsers cannot visit any other website, showing "connection refused". And the download would also fail (after the first part of this model finished, the next part cannot start and reports error). The log `~/AppData/Local/Ollama/server.log` shows `Error: listen tcp 127.0.0.1:11434: bind: Only one usage of each socket address (protocol/network address/port) is normally permitted.` I also used the `netstat`, which shows that port number up to 65535 is occupied by ollama. Is there any way to solve it? ### OS Windows ### GPU Nvidia ### CPU AMD ### Ollama version Win 0.3.12
GiteaMirror added the bugwindowsnetworking labels 2025-11-12 12:18:59 -06:00
GiteaMirror commented 2025-11-12 12:19:00 -06:00
Author
Owner
Copy Link

@dhiltgen commented on GitHub (Sep 26, 2024):

@TheStarAlight are you by any chance on a slow or flaky network where a lot of retries are happening? Perhaps we have a FD ~leak or delayed close in the retry logic and aren't cleaning up as we go.

@dhiltgen commented on GitHub (Sep 26, 2024): @TheStarAlight are you by any chance on a slow or flaky network where a lot of retries are happening? Perhaps we have a FD ~leak or delayed close in the retry logic and aren't cleaning up as we go.
GiteaMirror commented 2025-11-12 12:19:00 -06:00
Author
Owner
Copy Link

@TheStarAlight commented on GitHub (Sep 27, 2024):

@TheStarAlight are you by any chance on a slow or flaky network where a lot of retries are happening? Perhaps we have a FD ~leak or delayed close in the retry logic and aren't cleaning up as we go.

Yes you are right, network is not stable in China. Thank you for your attention! :D

@TheStarAlight commented on GitHub (Sep 27, 2024): > @TheStarAlight are you by any chance on a slow or flaky network where a lot of retries are happening? Perhaps we have a FD ~leak or delayed close in the retry logic and aren't cleaning up as we go. Yes you are right, network is not stable in China. Thank you for your attention! :D
GiteaMirror commented 2025-11-12 12:19:01 -06:00
Author
Owner
Copy Link

@dhiltgen commented on GitHub (Sep 30, 2024):

@TheStarAlight the fix will be in the next version (0.3.13) when that ships.

@dhiltgen commented on GitHub (Sep 30, 2024): @TheStarAlight the fix will be in the next version (0.3.13) when that ships.
GiteaMirror commented 2025-11-12 12:19:01 -06:00
Author
Owner
Copy Link

@SantaLaMuerte commented on GitHub (Oct 23, 2024):

@TheStarAlight исправление будет в следующей версии (0.3.13), когда она выйдет.

maybe add a flag --port to set a custom port for webui , cause port is used by ollama 11434 process itself?

also when you under vpn it might cause some missconfigs with localhost adress. but its another story

@SantaLaMuerte commented on GitHub (Oct 23, 2024): > @TheStarAlight исправление будет в следующей версии (0.3.13), когда она выйдет. maybe add a flag --port to set a custom port for webui , cause port is used by ollama 11434 process itself? also when you under vpn it might cause some missconfigs with localhost adress. but its another story
GiteaMirror commented 2025-11-12 12:19:01 -06:00
Author
Owner
Copy Link

@SantaLaMuerte commented on GitHub (Oct 23, 2024):

https://github.com/ollama/ollama/issues/5816
https://github.com/ollama/ollama/issues/3774

Get-Process ollama* | Stop-Process -Force | ollama serve
in one of issues - is founded solution.

Is a possible to build in an OpenUI inside (with WSL support or even just python packages) ? Kind of plug and play
it will be very helpfull to just install , open ui and start chat with files, internet and much more . Thanks for your product and work!

(easieast setup for me was install ollama , install webui via python and just use . thanks again)

@SantaLaMuerte commented on GitHub (Oct 23, 2024): https://github.com/ollama/ollama/issues/5816 https://github.com/ollama/ollama/issues/3774 `Get-Process ollama* | Stop-Process -Force | ollama serve ` in one of issues - is founded solution. Is a possible to build in an OpenUI inside (with WSL support or even just python packages) ? Kind of plug and play it will be very helpfull to just install , open ui and start chat with files, internet and much more . Thanks for your product and work! (easieast setup for me was install ollama , install webui via python and just use . thanks again)
GiteaMirror commented 2025-11-12 12:19:02 -06:00
Author
Owner
Copy Link

@TheStarAlight commented on GitHub (Oct 24, 2024):

#5816 #3774

Get-Process ollama* | Stop-Process -Force | ollama serve in one of issues - is founded solution.

Is a possible to build in an OpenUI inside (with WSL support or even just python packages) ? Kind of plug and play it will be very helpfull to just install , open ui and start chat with files, internet and much more . Thanks for your product and work!

(easieast setup for me was install ollama , install webui via python and just use . thanks again)

I think you should raise this question in open-webui. I'm also looking forward to a installable version. Using docker is difficult for freshman like me, and I usually encounter problems that the webui cannot start.

@TheStarAlight commented on GitHub (Oct 24, 2024): > #5816 #3774 > > `Get-Process ollama* | Stop-Process -Force | ollama serve ` in one of issues - is founded solution. > > Is a possible to build in an OpenUI inside (with WSL support or even just python packages) ? Kind of plug and play it will be very helpfull to just install , open ui and start chat with files, internet and much more . Thanks for your product and work! > > (easieast setup for me was install ollama , install webui via python and just use . thanks again) I think you should raise this question in [open-webui](https://github.com/open-webui/open-webui). I'm also looking forward to a installable version. Using docker is difficult for freshman like me, and I usually encounter problems that the webui cannot start.
GiteaMirror commented 2025-11-12 12:19:02 -06:00
Author
Owner
Copy Link

@SantaLaMuerte commented on GitHub (Oct 24, 2024):

#5816 #3774
Get-Process ollama* | Stop-Process -Force | ollama serve in one of issues - is founded solution.
Is a possible to build in an OpenUI inside (with WSL support or even just python packages) ? Kind of plug and play it will be very helpfull to just install , open ui and start chat with files, internet and much more . Thanks for your product and work!
(easieast setup for me was install ollama , install webui via python and just use . thanks again)

I think you should raise this question in open-webui. I'm also looking forward to a installable version. Using docker is difficult for freshman like me, and I usually encounter problems that the webui cannot start.

i anyway face a not able to found AMD 6700XT GPU , even with Hip drivers, amd latest (kind of Rocm for win) , but it shows that NOT DETECTED - so clear win installation is useless for me.. I see a gpu using ofk , but just a bit.

so the working way i think is anyway install it on WSL with real ROCM support or maybe docker))) (i dont like containers too)
i am one click guy, script kiddy - and cant even imagine what hell will go on simple people to made it all work )))
So claude or chatgpt is still the best way (20$ is 20$ ofk)

upd. found that https://medium.com/@danies.pahlevi/how-to-use-ollama-with-open-webui-with-rx-6700-xt-12gb-backend-in-windows-5aca4546e23f and ollama for amd repo .. but its again a костыли (crutches) ((

@SantaLaMuerte commented on GitHub (Oct 24, 2024): > > #5816 #3774 > > `Get-Process ollama* | Stop-Process -Force | ollama serve ` in one of issues - is founded solution. > > Is a possible to build in an OpenUI inside (with WSL support or even just python packages) ? Kind of plug and play it will be very helpfull to just install , open ui and start chat with files, internet and much more . Thanks for your product and work! > > (easieast setup for me was install ollama , install webui via python and just use . thanks again) > > I think you should raise this question in [open-webui](https://github.com/open-webui/open-webui). I'm also looking forward to a installable version. Using docker is difficult for freshman like me, and I usually encounter problems that the webui cannot start. i anyway face a not able to found AMD 6700XT GPU , even with Hip drivers, amd latest (kind of Rocm for win) , but it shows that NOT DETECTED - so clear win installation is useless for me.. I see a gpu using ofk , but just a bit. so the working way i think is anyway install it on WSL with real ROCM support or maybe docker))) (i dont like containers too) i am one click guy, script kiddy - and cant even imagine what hell will go on *simple* people to made it all work ))) So claude or chatgpt is still the best way (20$ is 20$ ofk) upd. found that https://medium.com/@danies.pahlevi/how-to-use-ollama-with-open-webui-with-rx-6700-xt-12gb-backend-in-windows-5aca4546e23f and ollama for amd repo .. but its again a костыли (crutches) ((
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
client2
cloud
compatibility
context-length
create
docker
documentation
embeddings
engine
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
install
integration
intel
js
linux
macos
memory
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
windows
wsl
No Label bug 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-ollama#4416
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?