95 Replies
Pod ID?
Please send pod id
Btw is it possible that the pod is down and it has no notice?
8wznoz1csrmful
We check pods every minute. So could be before the minute, the pod was checked
log
Your hf isn't authenticated
And im not sure about this error cannot remove 'Notebooks.txt': No such file or directory
Might be from your template
hf shouldnt even b connected
it throws you are not authorized
are you using some costum docker image?
What do you mean it shouldnt be connected
Isk how , I literally delete my pods when im not using them
i havnt logged into it
nope
fast stable defusion template
@TheLastBen did your HF mayby got hacked?
just closed and deleted pop and tried to start a ned one same thing
Now i'm getting a cloudflare error
Yeah you should if you want to use that private dataset, oh but this is totally unrelated to the downed pod network right
@nerdylive it should not require auth for that one it's looks like that dataset it's deleted from HF servers
oh its storing a notebook file
ic
good thing there are some forks
it's deleted or private repo its not there anymore yea
Runpod decided to discontinue supporting fast stable diffusion template
I never heard that info?
huh why
@TheLastBen where is that info comes from?
kinda want to understand what happened
Cause it kinda not make sense
Lots of people use that template so makes no sense to me either.
I was informed by email
Would you be still intresting on letting it be as community template.
I would be happy to still continue it as I also liked template.
Sure its not the dreambooth endpoint rather than the pod template? RunPod isn't doing hosted endpoints anymore.
GitHub
GitHub - kodxana/madiator-docker-runpod: RunPod Docker Containers f...
RunPod Docker Containers for RunPod. Contribute to kodxana/madiator-docker-runpod development by creating an account on GitHub.
Template is still there so someone probably got their wires crossed and meant that the dreambooth endpoint is getting discontinued. I don't see any reason to discontinue the pod template.
@Marcus got offcial info and yes it got discontinued
Wow, I wonder why
The template should be delisted in that case, isn't it?
So for those of us using that template with network storage with many models and such... what is the best template to change to?
the default Pytorch...
or custom template with your code or script
and then start A1111 how? (sorry always done it via the notebook)
Hmm call the start script of a1111
just use the a1111 template for the easiest way i guess
probably best thing would to be migrate to other template
what why
as the template is getting removed
ah ic
why is it getting removed?
I mean unless @TheLastBen would allow me to continue and adopt the template as community one
and what template is suggested to change to, for those of us not python experts
Kinda asking as it's his work so I respect the decision.
use ashleyk's a1111 template probably
👍 will try that.... just hoping it doesnt mess up the network storage... really wish runpod had a way to copy network storage
Why can't TheLastBen just maintain it as a community template instead of a RunPod official one? Or is he not interested in doing so.
would be nice... but quite curious the runpod reasoning here...?
I think it's some early adopter agrements but cant tell much
or it's just case that there are a lot more templates nowdays
Too many actually
Especially broken ones
RunPod should focus more on removing broken templates causing issues than taking away things that work
planning something to help solve the case
for now Explorer page is not being much curated
Yeah its like pulling the handle of a slot machine and hoping you get lucky
most ppl rely on external link to get a good one
hahah
I mean #🧩｜templates is best source for good templates
yeah like that probably
Templates need ratings so people can find the best ones
Almost like an app store
I agree
good idea❤️
Anyway, thanks for your valuable contributions TheLastBen, RunPod would not be where it is today without people like you who helped make it what it is.
is it just me or doesn't any other template work ? or are they that much slower to serup /
setup
They are slower thats why it was called "Fast Stable Diffusion"
The others have much larger docker images which take a long time to pull, and also not cached on all machines since they are Community and not Official ones.
lol
even the offical sd is broken
Official SD as in a1111? its just out of date not broken isn't it?
so the fastest loading and most usable SD template is getting the boot....hmmmmm
You can use official pytorch templates, they load fast, but then you have to install everything yourself, so there are pros and cons.
A lot of people used that Fast SD template so it honestly makes no sense to me why RunPod are getting rid of it.
yeah , i don't get it
And worst of all, is that they don't delist it, so people spin up pods and waste their money because it can't start.
i was using it , went for dinner and come back and couldnt launch it again
how to start webui with pytorch?
Add the port(s) to your pod after installing the webui.
RunPod Stable Diffusion
is broken
thats it?
add the ports and restart pod?
Broken how?
Pod will restart by itself when you save the ports if I can remember correctly.
error 504
bad gateway
yes but it doesnt start webui
cloudflare
sorry 502
What do the logs say?
502 usually happens when its syncing to /workspace
like 9mins
Install script for stable-diffusion + Web UI
Tested on Debian 11 (Bullseye), Fedora 34+ and openSUSE Leap 15.4 or newer.
################################################################
################################################################
ERROR: This script must not be launched as root, aborting...
################################################################
Takes about 15+ mins
15 mins to start , this is doing my hed in i'll try again in the morning
Where did you see this?
trying to instal A1111
Install it where? Pytorch template?
if you guys figure something out can you tag me and let me know please
You're probably using network storage, its slow to sync to network storage, no way around that really.
yeah I have no idea what I am doing, this is is freaking frustrating... the fast SD just worked and now nothing works, spent 40-50 a week on runpod now about to give up
But you should be root on pytorch template, so you shouldn't get that issue
This template has latest A1111 webui and some cool extensions preinstalled
So I have run into this before with ashleyk templates, they all get stuck on syncing main venv to workspace...
if it ever finishes, its usually 20-30 minutes
yep
They finish, there is no "if they ever finish", network storage is just slow.
And they are not stuck at all, it says wait so you have to follow the instructions and wait.
paying to wait, awesome
such improvement
It only has to sync the first time, once its synced it doesn't have to sync again, and there is also a
DISABLE_SYNC environment variable that can be set to disable syncing.So sync just this once, then add that anytime I go to deploy a pod so it doesnt sync again?
it's slow on network storages