Zero GPUs all day.
Hello, it's been rough today and there have been no GPUs available for 5090 all day on EU-No-1.
Is there an outage we should know about, or is this just high usage?
9 Replies
Same issue
I'm deploying a new pod, I can't lose the productivity.
I can’t deploy a new pod because each of my existing servers already has over 150 GB of essential files. Re-setting everything, re-configuring my API, and updating all server connections would take a lot of time and reduce productivity.
Moreover, this isn’t the first time we’ve experienced this issue, which makes the situation even more frustrating.
This ongoing problem honestly makes me reconsider continuing with RunPod :/
I'm brand new with Runpod. I hope this doesn't remain an issue.
Unknown User•2w ago
Message Not Public
Sign In & Join Server To View
I'm not trying to refuse it, I was just not aware of how necessary it would be
I'm a bit confused about how I'm going to do this, but I've built the volume itself. I am trying to discern how I can build my pod again, I'm not looking forward to spending hours installing dependencies again.
I have the volume itself created... Now I just need to figure out how to redeploy everything I already have on my other pod.
I guess I'm honestly confused about this networked volume, like I can run any pod with this particular volume? How does this help with GPUs being unavailable? I still have to have a GPU available, and does it not still have to be on the same server? Is separating them going to decrease overall efficiency, as well?
Each DC has its own set of GPUs available. Each network volume can be used only in the DC where it was created. May be you created the volume at that DC which doesn't have the GPU you need.
Unknown User•2w ago
Message Not Public
Sign In & Join Server To View
I didn't realize, so instead I just spent a few hours rebuilding the models and such from scratch.
But every time I do this, my outputs are so different. It's pretty strange.