Hi everyone. I'm usually over in the cloud ops section, but I've been using Suno a bit for some creative breaks. I'm running into a big slowdown today and wanted to see if it's just me.
My generations are all taking over 5 minutes to complete, and a few just timed out. I checked my own AWS setup (no issues there) and my network seems fine. Is anyone else seeing major delays right now? Wondering if there are known server issues. 😅
I'm used to troubleshooting with Terraform, but here I'm a bit lost. Just looking for a quick sanity check. Thanks.
Hi user316, welcome over from cloud ops. You're definitely not alone in seeing this today. A few other members have mentioned similar timeouts in the general chat thread, starting a couple hours ago. It does seem to point to a wider service issue rather than local infrastructure.
Given your background, the fact you've already ruled out your own AWS environment and network is super helpful context. That's the exact kind of first-step isolation we encourage. For now, it's likely a waiting game until their team resolves things on their end, which can be frustrating when you're just looking for a creative break.
Have you noticed if it's affecting all generation types equally, or is it more pronounced with certain settings, like longer prompts? That kind of data can sometimes hint at the scale of the backend problem. Hope it clears up soon for you.
Stay curious.
Solid point about checking generation types. It's a good way to triage if it's a compute bottleneck or something else in the pipeline.
If I were them, I'd be looking at the cost of scaling up whatever instance type they're using for inference. Those spot fleets get pricey when demand spikes, and you can bet their margins just evaporated for the day. A 5-minute queue suggests their auto-scaling groups are lagging way behind, or their load balancer routing is a mess.
Show me the bill