invoke-ai
[bug]: Suddenly stuck in a VAE install loop while unable to find pytorch model? (no longer boots)
Is there an existing issue for this?
- [X] I have searched the existing issues
OS
Linux
GPU
cuda
VRAM
No response
What version did you experience this issue on?
3.0.0.+a0
What happened?
Sorry it feels like I am posting here a lot, but I just keep waking up to the whole setup deciding it wants to be different every day lol.
Been working fine for past few days (albeit I still think the VAE is broken on facial stuff and negative prompting) but I suddenly launched last night to find that it was saying pytorch models aren't found, and it trying to re-install somewhat, then pretend everything is fine, then launch webui... then it doesn't do anything. It just kinda sits there like "job well done" and WebUI won't work, etc.
I thought it was just downloading a new push update (as my setup is set to always follow main branch automatically, which I am starting to think.. is not the move lol)
Here's the log:
`[21-05-2023 21:18:02]::[InvokeAI]::INFO --> Patchmatch initialized [21-05-2023 21:18:03]::[InvokeAI]::INFO --> InvokeAI, version 3.0.0+a0 [21-05-2023 21:18:03]::[InvokeAI]::INFO --> InvokeAI runtime directory is "/home/invokeuser/userfiles" [21-05-2023 21:18:03]::[InvokeAI]::INFO --> GFPGAN Initialized [21-05-2023 21:18:03]::[InvokeAI]::INFO --> CodeFormer Initialized [21-05-2023 21:18:03]::[InvokeAI]::INFO --> Face restoration initialized [21-05-2023 21:19:17]::[InvokeAI]::INFO --> Patchmatch initialized [21-05-2023 21:19:17]::[InvokeAI]::INFO --> InvokeAI, version 3.0.0+a0 [21-05-2023 21:19:17]::[InvokeAI]::INFO --> InvokeAI runtime directory is "/home/invokeuser/userfiles" [21-05-2023 21:19:17]::[InvokeAI]::INFO --> GFPGAN Initialized [21-05-2023 21:19:17]::[InvokeAI]::INFO --> CodeFormer Initialized [21-05-2023 21:19:17]::[InvokeAI]::INFO --> Face restoration initialized [21-05-2023 21:19:48]::[InvokeAI]::INFO --> Patchmatch initialized [21-05-2023 21:19:48]::[InvokeAI]::INFO --> InvokeAI, version 3.0.0+a0 [21-05-2023 21:19:48]::[InvokeAI]::INFO --> InvokeAI runtime directory is "/home/invokeuser/userfiles" [21-05-2023 21:19:48]::[InvokeAI]::INFO --> GFPGAN Initialized [21-05-2023 21:19:48]::[InvokeAI]::INFO --> CodeFormer Initialized [21-05-2023 21:19:48]::[InvokeAI]::INFO --> Face restoration initialized From https://github.com/invoke-ai/InvokeAI ff0e79fa..650d69ef main -> origin/main 02d2cbcd..23abaae9 feat/images-service -> origin/feat/images-service [22-05-2023 07:45:17]::[InvokeAI]::INFO --> Patchmatch initialized [22-05-2023 07:45:17]::[InvokeAI]::INFO --> InvokeAI, version 3.0.0+a0 [22-05-2023 07:45:17]::[InvokeAI]::INFO --> InvokeAI runtime directory is "/home/invokeuser/userfiles" [22-05-2023 07:45:17]::[InvokeAI]::INFO --> GFPGAN Initialized [22-05-2023 07:45:17]::[InvokeAI]::INFO --> CodeFormer Initialized [22-05-2023 07:45:17]::[InvokeAI]::INFO --> Face restoration initialized [22-05-2023 07:55:55]::[InvokeAI]::INFO --> Patchmatch initialized [22-05-2023 07:55:55]::[InvokeAI]::INFO --> InvokeAI, version 3.0.0+a0 [22-05-2023 07:55:55]::[InvokeAI]::INFO --> InvokeAI runtime directory is "/home/invokeuser/userfiles" [22-05-2023 07:55:55]::[InvokeAI]::INFO --> GFPGAN Initialized [22-05-2023 07:55:55]::[InvokeAI]::INFO --> CodeFormer Initialized [22-05-2023 07:55:55]::[InvokeAI]::INFO --> Face restoration initialized From https://github.com/invoke-ai/InvokeAI 0ce628b2..fab5df93 v2.3 -> origin/v2.3 [22-05-2023 07:57:49]::[InvokeAI]::INFO --> Patchmatch initialized [22-05-2023 07:57:49]::[InvokeAI]::INFO --> InvokeAI, version 3.0.0+a0 [22-05-2023 07:57:49]::[InvokeAI]::INFO --> InvokeAI runtime directory is "/home/invokeuser/userfiles" [22-05-2023 07:57:49]::[InvokeAI]::INFO --> GFPGAN Initialized [22-05-2023 07:57:49]::[InvokeAI]::INFO --> CodeFormer Initialized [22-05-2023 07:57:49]::[InvokeAI]::INFO --> Face restoration initialized Downloading sd-inpainting-1.5: Downloading stable-diffusion-2.1: ** models.yaml exists. Renaming to models.yaml.orig Successfully created new configuration file /home/invokeuser/userfiles/configs/models.yaml
** INVOKEAI INSTALLATION SUCCESSFUL ** If you installed manually from source or with 'pip install': activate the virtual environment then run one of the following commands to start InvokeAI.
Web UI: invokeai-web
Command-line client: invokeai
If you installed using an installation script, run: /home/invokeuser/userfiles/invoke.sh
Add the '--help' argument to see all of the command-line switches available for use.
Loading InvokeAI WebUI..... invoke> Checking if The Git Repo Has Changed.... Local Files Are Up to Date Loading InvokeAI WebUI..... invoke> Checking if The Git Repo Has Changed.... Local Files Are Up to Date Loading InvokeAI WebUI..... invoke> Checking if The Git Repo Has Changed.... Updates Found, Updating the local Files.... Updating ff0e79fa..650d69ef Fast-forward invokeai/frontend/web/src/app/components/InvokeAIUI.tsx | 11 +++++++++-- .../src/features/gallery/components/CurrentImageButtons.tsx | 8 +++++++- .../web/src/features/gallery/components/HoverableImage.tsx | 8 +++++++- invokeai/frontend/web/src/features/gallery/store/actions.ts | 4 ++++ 4 files changed, 27 insertions(+), 4 deletions(-) Loading InvokeAI WebUI..... invoke> Checking if The Git Repo Has Changed.... Local Files Are Up to Date Loading InvokeAI WebUI..... invoke> Checking if The Git Repo Has Changed.... Local Files Are Up to Date Loading InvokeAI WebUI..... invoke> `
Screenshots
No response
Additional context
No response
Contact Details
Following and updating regularly to Main, which is in an active migration/development, is going to cause you a lot of this.
Will let others opine on current issue you're facing, but it's not clear to me whether you're initializing using the new Nodes backend or the legacy web server (which was recently removed)
I don't mind continuing to post bugs/have issues if it means it actually helps you guys tweak stuff and I am somewhat helping the community in being vocal, I just don't want to feel like I am the welfare guy expecting everyone to solve all my problems all the time.
I would guess I am using the legacy web servers? Based on this note from the docker creator:
"Latest docker changes (4 May 23): just uploaded a quick fix to docker hub, locked the current main branch to the pre-nodes tag till main migration to nodes is done, however if there is files under /invokeai/invokeai/ they will need to be deleted before starting the docker so it can pull the pre-nodes tag"
The legacy web server has been removed. You'll need to launch the new web server with invokeai-web (not invokeai --web) from the command line:
source ~/invokeai/.venv/bin/activate # linux
C:\YourUserName\invokeai\.venv\Scripts\activate # windows
invokeai-web
Note that the web script currently requires you to be in the root directory of the repo.
We've basically thrown all the pieces up in the air and are trying to put them back together again, so various bits of the install and configure process are not working.
I don't mind continuing to post bugs/have issues if it means it actually helps you guys tweak stuff and I am somewhat helping the community in being vocal, I just don't want to feel like I am the welfare guy expecting everyone to solve all my problems all the time.
You're good! So long as you're ok with a "not expected to work", report away. Definitely helps keep us on our toes :)