Help! Why do some of my workers report insufficient space when pulling images?
Hello everyone, I have a question for you. As shown in the picture, some of my workers will report insufficient space, but most of them will not.
9 Replies
I'm not sure what caused this problem.
1. When creating the endpoint, I chose 100g for the container disk size, but my image only has 15g.
2. I found the previous history and found that it might be a problem with the cuda version allowed when creating the endpoint. I'm currently trying
@hammer
Escalated To Zendesk
The thread has been escalated to Zendesk!
try opening a support ticket
The image you sent says no space on device when pulling docker image which happens on the host, not your container
so its probably a host problem (so runpod's problem) and not your image's fault
I understand that multiple workers are running on the same endpoint (that is, in the same machine?). It may be because I chose a large disk storage.
No it isnt in the same machine but maybe because you create a large disk
But runpod should account for that and make the container use a host with that amount of storage availiable instead of it failing like this
Probably, if you want you can check on the rest api or graphql
I think the way to solve this is to let support check
guys!
This is a problem with the runpod region (because I used a net volume)
Switching to another region will solve the problem
The runpod team is working on fixing this problem
@riverfog7 @Jason
Thank you guys!
I see
Your welcome!