Finish / Start of AutoInst on Runpod not possible
Steed9746 opened this issue · 7 comments
Hi, and first of all, thank you a lot for your work and explanation!
Worked at Runpod, RTX A6000 like you described in your youtube auto install video.
And every think worked fine till i executed:
cd /workspace/stable-diffusion-xl-demo
source venv/bin/activate
PYTORCH_CUDA_ALLOC_CONF=max_split_size_mb:256 SHARE=true ENABLE_REFINER=true python app7.py
Here i got several warnings as a feedback (see below what i reproduced when started again)
Interestingly it gives me a 127. ip to start.
- when i click the 127, there is nothing - as i expected
- wenn i start the web terminal and connect to web terminal, i get only this: "root@38f87f9b1020:/workspace#" on a black page.
- When i click connect to HTTP... Port3001, i get the 502 Page
Thanks for the support
El
root@38f87f9b1020:/workspace/stable-diffusion-xl-demo# cd /workspace/stable-diffusion-xl-demo
source venv/bin/activate
PYTORCH_CUDA_ALLOC_CONF=max_split_size_mb:256 SHARE=true ENABLE_REFINER=true python app7.py
Loading model /workspace/stable-diffusion-xl-base-0.9
Loading pipeline components...: 29%|█████████████████████████████████████▋ | 2/7 [0Loading pipeline components...: 43%|████████████████████████████████████████████████████████▌ | 3/7 [0Loading pipeline components...: 86%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████▏ | 6/7 [0Loading pipeline components...: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 7/7 [00:32<00:00, 4.62s/it]
Loading model /workspace/stable-diffusion-xl-refiner-0.9
Loading pipeline components...: 100%|█| 5/5 [00:11<00:00,
/workspace/stable-diffusion-xl-demo/venv/lib/python3.10/site-packages/gradio/deprecation.py:43: UserWarning: You have unused kwarg parameters in Blocks, please remove them: {'timeout': 300}
warnings.warn(
/workspace/stable-diffusion-xl-demo/venv/lib/python3.10/site-packages/gradio/layouts.py:75: UserWarning: mobile_collapse is no longer supported.
warnings.warn("mobile_collapse is no longer supported.")
/workspace/stable-diffusion-xl-demo/venv/lib/python3.10/site-packages/gradio/components.py:198: UserWarning: 'rounded' styling is no longer supported. To round adjacent components together, place them in a Column(variant='box').
warnings.warn(
/workspace/stable-diffusion-xl-demo/venv/lib/python3.10/site-packages/gradio/components.py:216: UserWarning: 'border' styling is no longer supported. To place adjacent components in a shared border, place them in a Column(variant='box').
warnings.warn(
/workspace/stable-diffusion-xl-demo/venv/lib/python3.10/site-packages/gradio/components.py:207: UserWarning: 'margin' styling is no longer supported. To place adjacent components together without margin, place them in a Column(variant='box').
warnings.warn(
/workspace/stable-diffusion-xl-demo/venv/lib/python3.10/site-packages/gradio/deprecation.py:43: UserWarning: You have unused kwarg parameters in Slider, please remove them: {'enabled': True}
warnings.warn(
Running on local URL: http://127.0.0.1:7860
@El-Sonador sorry for delay
i am still testing will let you know soon
@El-Sonador
I have updated the auto installer
Make sure that your container disk size is at least 30 GB
volume disk size is not anymore important
use latest script from here
https://www.patreon.com/posts/85845581
it will give you gradio share link after run use it
Hi Furkan,
thanks a lot for the fast support ! ! ! Thats extraordinary!
Testing it at the moment.
What does "volume disk size is not anymore important" mean? Can i stick with 75?
This would dramatically decrease the standby price and would be highly appreciated.
Hi Furkan, thanks a lot for the fast support ! ! ! Thats extraordinary! Testing it at the moment.
What does "volume disk size is not anymore important" mean? Can i stick with 75? This would dramatically decrease the standby price and would be highly appreciated.
yes you can even use 50 gb
it will load models into container disk with minimal amount
Wow, great improvement! 👍
But actually i will stay with 1.5. After testing XL, it feels like a very limited Demo version of 1.5. There are so much functions missing (or more realistically, just did not found them yet). 😄
Wow, great improvement! 👍 But actually i will stay with 1.5. After testing XL, it feels like a very limited Demo version of 1.5. There are so much functions missing (or more realistically, just did not found them yet). 😄
ye it is getting better over time
:) hope so. Will close this, sine you solved it. thank you.