RunPod

We're a community of enthusiasts, engineers, and enterprises, all sharing insights on AI, Machine Learning and GPUs!

Join

RunPod

We're a community of enthusiasts, engineers, and enterprises, all sharing insights on AI, Machine Learning and GPUs!

Join

⚡｜serverless

⛅｜pods-clusters

jphipps

3/5/2025

Network and Local Storage Performance

Hi, we are noticing very slow performance loading in our model on our Pods in the IS region. We are also noticing a very slow sequential read time when we copy the same model into local storage. The model loading takes about 10x as much time as it did for us on a different network. When we compare the sequential read time, we see about a 3x increase in time on Runpod. Local storage is about 5s faster than network storage. our old network read ```13550863+1 records in 13550863+1 records out 6938042106 bytes (6.9 GB, 6.5 GiB) copied, 7.09554 s, 978 MB/s...

franrappazzini

3/5/2025

Run own CPU

Hey! I have the following machine. Would I run this machine on RunPod to gain rewards? Thanks!

Platinum

3/5/2025

2-3 hours for Pod to load OneTrainer and Flux model

See images, marked with red is how long it took to load. I really want to use cloud for Flux LoRa Training with OneTrainer and using the OneTrainer cli Template with flux model it took 2-3 hours to prepare before training can begin. Is this correct that I should wait 3 hours before training begins for a Flux LoRa? If I have done something wrong what is it and how do I correct it?...

Ahmad Alfan

3/5/2025

Pod SSH Connection Slow and Failing in EU-SE-1

I'm using "SSH over exposed TCP", either SSH connection from VS Code or Terminal is very slow, from terminal I got connected but occasionally got these logs: ``` channel 22: open failed: connect failed: open failed channel 24: open failed: connect failed: open failed channel 26: open failed: connect failed: open failed...

jphipps

3/4/2025

4090 Power capped

Hi, I was testing an inference job on a 4090 pod. I noticed it was running very slowly. When I checked the nvidia logs, I noticed a "sw power cap" message when it got to about 1/3 of the Power (450W). How do we get full performance of our 4090 GPU?

Olbanets

3/4/2025

Network volume for GPU and CPU

I'd like to try two GPU with my big dataset so I was looking for a DC with the both GPU and found one (US-KS-2). But I also found that network volumes aren't available for CPU at that DC. Am I right? Is there a chance to get this option in the near future?...

Solution:

CPU pods are only at EU-RO-1 and EUR-IS-1

pavanmantha

3/3/2025

Unable to run any docker image in runpod instance

I have setup the instance successfully, installed the docker successfully but unable to run any image in the instance. It always fails with below error. ```root@6f45062b53d4:~# docker ps CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES root@6f45062b53d4:~# docker run hello-world Unable to find image 'hello-world:latest' locally...

sayuto

3/3/2025

HOw can I open port for my pod and run it as a api endpoint

I can not ping to my endpoint I can not run nginx in my pod How can I open the port? I created it with RunPodPyTorch 2.4 and global network enabled...

DIRECTcut ▲

3/2/2025

[runpodctl] Error creating a pod with a network volume attached

Hi. I am getting an error trying to create a pod with an existing network volume attached. The erroring command: ``` runpodctl create pod --secureCloud --networkVolumeId=="214w8k0zq1" volumePath="/workspace" --gpuType="NVIDIA GeForce RTX 4090" --imageName="ubuntu:latest" Error: Something went wrong. Please try again later or contact support....

chronoz22

3/2/2025

Mimicking UI with API issues

Hellow! Im trying to use the template byecho/simpletuner-image:latest via the API. When i deploy via UI it all works fine and i get a public IP and port that i can ssh/scp into. ...

Fluxz

3/1/2025

Pod Not Starting

Resolving github.com (github.com)... failed: Temporary failure in name resolution. wget: unable to resolve host address 'github.com' This is the error I am getting ^^^^ Am I doing something wrong?...

Aiwolf

3/1/2025

How to copy current settings into a template?

I basically always use the same settings and have to edit the template each time. How can I just copy the settings I have for my pods and save it as a template?

Jaunty

3/1/2025

Pod with multiple gpus (rtx 4090)

When I am starting runpod/pytorch:2.4.0-py3.11-cuda12.4.1-devel-ubuntu22.04 with multiple gpus (rtx 4090) in my simple .cu file I am just trying to get cudaGetDeviceCount and getting 999 unknown error. when running code on instance with one gpu all is working fine. in nvidia-smi all gpus are visible. maybe i am missing anything obvious in pod setup?

riverfog7

3/1/2025

CUDA error in community 4090x4 pod

https://github.com/BAI-Yeqi/PyTorch-Verification using this script gives this error ...

Rob Guerrero

3/1/2025

How to connect using WinSCP to community pod?

I created a pod with community option and clicked the public IP. When I check the connect button on web UI I dont see IP address. When I use runpodctl get pod <pod id> -a it does return pod info with IP address = 100.65.10.109:60679->19123 (prv,http),100.65.10.109:60678->8020 (prv,http) . I assume "prv" means private IP? where is the public IP? Using WinSCP (latest version) I picked SFTP file protocol and enter Ip address = 100.65.10.109 under hostname and port= 22 and set the private key in advanced button but it cant find host. I tried "ping 100.65.10.109" and it cant find it so I assume that IP is private? Or is there a way to just use the pod ID as the hostname? Thanks in advance...

Milad

2/28/2025

IS datacenters slow connectivity

We are seeing high latency in IS data centers. Both on pods and serverless workers. Anyone else?

justblake3911

2/28/2025

Experiencing some POD connection issues in CZ

Anyone have updates on any connecting issues reported in CZ data center for Pods? Thanks,

TJS104

2/28/2025

Can't access Checkpoints folder via JupyterLabs

Hi all, I have an installation of Comfyui on my pod. When I'm in JupyterLabs, if I double-click on any of the folders over in the left-hand pane I can enter that folder. However, when I double-click on the comfyui 'Checkpoints' folder nothing happens! I can even right-click the folder and choose 'Open', but still nothing. The folder is: workspace/ComfyUI/models/checkpoints. Now, not only does this mean that I can't actually see what's in that folder, it also means, as far as I can tell, that I've no way of deleting anything in that folder (which I need to do, to clear some unwanted checkpoints). Is this a bug in JupyterLab, or a bug within the pod? It's driving me crazy. If anyone knows of a solution, or a workaround, I would be delighted to hear it. Best to all, T...

Patricky7777

2/28/2025

Pod stopped and I cannot re access

Hi 🙂 Forgive my noobness. Pod stopped as I didn't notice funds running out. I re-uped the funds but now cannot connect to any GPUs. I need to access the volume. Would also like to continue using the setup if possible. How can I? - What's the problem? Thanks in advance. Sorry for any dumbness....

DinhSon

2/27/2025

Pod easily get OOM!

I am using an 8xA40 instance. Pod id: k3urxcxexkj989 Even though I do not run any heavy tasks, just unzip a file and upload some data to the pod using scp commands, the pod frequently got OOM issues. My pod has ~375GB of RAM, and I don't think my process caused the problem. Could you check out the issue? Thanks...

Previous Next

Gaming

Programming

RunPod

We're a community of enthusiasts, engineers, and enterprises, all sharing insights on AI, Machine Learning and GPUs!

RunPod

We're a community of enthusiasts, engineers, and enterprises, all sharing insights on AI, Machine Learning and GPUs!