Missing port buttons and Unable to “start web terminal” on Ultimate Template
Hello,
I've been trying to install @ashleyk Stable Diffusion Kohya_ss ComfyUI Ultimate 3.12.1 all afternoon with no success. Typically, I can be up and running on a new pod in ~10-20 minutes; today, it was taking much longer. The log would often hang for ~10-15 minutes during each step to Sync (kohya, comfyui, Automatic) Venv. Once the installation was complete, I clicked the connect button to start Auto1111, comfyUI, Jupyter, etc., but the UI service buttons were missing. I tried installing this container on an a40, a6000, and a5000. I burned through $2 just on unsuccessful installation time. Today is the first time anything like this has occurred. I genuinely appreciate your work @ashleyk . I'd love to learn how to create or tweak the current container.
31 Replies
It's normal for the syncing to take a long time to compete especially with a network volume, you have to wait for it to complete. This is not an issue .
View the container logs to see the progress of the syncing.
As far as I can tell, the entire installation process completed as seen in attached photos. @ashleyk
Not sure why the ports are not showing up. Must be some RunPod issue @flash-singh . You may want to try using Communituy Cloud if you're not using Network Storage, the Community Disk is at least 20 times faster than the horrific speeds you get from Network Storage or using regions that have network storage available.
The ports just popped up for a second, then disappeared again.
Probably some kind of RunPod issue, nothing I can control unfortunately.
Thank you for your reply and work!
oof, now the pod is stuck on "waiting for logs" and now a "critical error."
Give the pod id to RunPod so that they can check it out for you.
It's been a long day and the pod has already been tossed in the trash.
thanks again for your help. I'll try again tomorrow and reach out to RP if I encounter the same issue.
Same issue as yesterday on 2 attempts of fresh install & default settings of "Stable Diffusion Kohya_ss ComfyUI Ultimate 3.12.1" on Secure Cloud a5000 SE server. I'm not using network storage. Ports are not showing up, "Start Web Terminal" does nothing.
Could this be due to new feature "Runpod Secret" under Edit Pod/Environmental Variables? @ashleyk
Container logs always end with "sshd: no hostkeys available -- exiting."
Same experience on Community Cloud.
There is no issue with the template, many people are using it. You are likely not waiting for the applications to sync to /workspace. Check the logs.
What should I be looking for in the logs? The log always ends with "sshd: no hostkeys available -- exiting." And the ports never show up.
I’m not doing anything differently than before. I hope I’m not sounding accusatory, I just have no idea why it’s not working as usual. Thank you again for the reply.
The attached photo with the log appears to show that everything has started up correctly right?
Ok will have a look later, not sure what causes this, I have not been able to reproduce it, can you try selecting a region that does not have network storage?
I can try, but I do not know how to select a region that does not have network storage. I have never added a Network Volume (with "Select Network Volume") to any of these install attempts.
The ports were appearing during the Syncing to Venv phase, but when "All services have been started" the ports vanish.
_
Ports appearing during Venv sync phase
_
Ports Vanish after "All services have been started"
After "All services have been started," The ports appeared two times for ~15 seconds then the ports vanished and never returned.
This indicator is in constant loop of going to 34 seconds then back to 7 seconds.
System Log is 80 lines of "start container." Every 30 seconds a new "start container" line is created.
Fyi, This is how I launch a new Template:
Explore > Stable Diffusion Kohya_ss ComfyUI Ultimate > Deploy to GPU Cloud > Select: Secure Cloud, Any Region, > Deploy 1x RTX A5000 > Default settings Continue > Deploy
Thanks again!
Please let me know if I can provide any other information that might help. I hope I have been selecting a region without network storage. @ashleyk
All of these have network storage.
I tried one in US-TX, it was looking promising, but system log is once again printing 'start container' over and over again.
Ports have vanished again.
Would a specific browser, a VPN, clearing cache on my end possibly help? @ashleyk
That start container thing is probably an issue with the host machine, not sure why there are so many broken pods lately. @Philllis I think the original issue you had may have been because the start.sh script had
set -e at the top which would have crashed the script on error, so I have removed it now and currently building a new version of the template. I'll let you know when its ready. In the meantime I DM'd you a credit code for the inconvenience.
Don't think so.Ok! Thank you kindly for your help and eyes on the issue I've been encountering. I hope I wasn't too bothersome! 😅
Yea, i've also noticed more host machine issues over this past month.
It was unnecessary, but thank you kindly for the credit code. Your template has been a great help to me as I learn and create with these new tools.
@Philllis I've just finished pushing version 3.12.2 which should hopefully resolve the issue if you want to give it a try.
Up and running! Thank you! 🤘
@ashleyk
Awesome, glad its working now, apologies for the inconvenience.
this usually means the container keeps stopping, and we start it again, this will result in ports appearing and disapearing
It was
set -e in the start.sh script that caused the entire start script to crash when something failed, so yeah. Its been fixed now.
I am having an issue I can't figure out how to work around it. I am new to RunPod so please excuse the lack of knowledge at this point. I have my pod running and my SSH pub key configured. But when I attempt to ssh into it using my private key, it seems to connect...but then I get the error "Error response from daemon: Container 7b7a3790f1500c544348d2c6e09c286ee3fe3849adcb241ac54bceb3c518619f is not running"....even though I can see the pod running
Clicking on the "Start Web Terminal" on the UI doesnt do anything either....
Please don't ask the same question u posted in a different question / someone else's question that is irrelevant 🙂 Ill answer the question on the other one u posted, but no need to double post it
Apologies @justin my bad... Thanks for your help