Hey everyone! Ran into a weird one today and hoping someone has seen this before.
I was setting up a new CloudFront distribution to serve some marketing assets (landing pages, PDFs, etc.) from an S3 origin. The config was pretty standard—no fancy Lambda@Edge or anything. Hit "Create," and it's been stuck in the **"In Progress"** state for over four hours now. Usually, this takes 5-10 minutes, right?
Here's what I've already checked:
- The S3 bucket permissions (bucket policy and CORS) look correct.
- No recent changes to my AWS account limits.
- The distribution *is* listed in the console, but the state hasn't budged from "In Progress."
Has anyone else hit this kind of delay recently? I'm wondering if:
* There's a regional service issue they haven't posted about.
* Something specific in my config (like a mis-typed header) could cause a silent hang instead of an error.
* I should just delete and rebuild, or if that'll make it worse.
Would love any tips if you've been through this! I need this CDN live for a campaign sync tomorrow 😅
Cheers,
Anna
Keep it simple.
Hey Anna, had a similar scare last month! Mine was stuck for about three hours. Turns out, in my case, it was a weird thing with the SSL certificate I'd chosen. I was using an ACM cert from a different region (us-east-
Did you happen to check the AWS Service Health Dashboard? Sometimes they have delays that aren't full-blown "issues" yet. The console might not show an error even if their backend provisioning is bogged down.
If it's been four hours, I'd probably lean towards deleting and rebuilding it now, since you need it for tomorrow. That's what I ended up doing and the second one deployed in minutes. Just make sure to note your exact settings first, obviously.
Quick question, was your S3 bucket name formatted in a weird way, like with periods? I heard that can cause issues sometimes, though usually it errors out upfront.
Learning by breaking
Ugh, that sounds incredibly frustrating, especially with a campaign deadline looming! Your checks are spot-on, and I agree the regional service issue is a real possibility they don't always broadcast.
> I'm wondering if... something specific in my config (like a mis-typed header) could cause a silent hang instead of an error.
This was my first thought too. I've seen a malformed `Cache-Control` header setting in the origin config cause a similar silent stall, where it just spins forever. Could you double-check any custom origin response headers you added? It's a long shot, but sometimes the error messaging just fails.
Since it's been hours, I'm leaning towards user162's suggestion to rebuild, as painful as it is. Before you do, maybe quickly create a dummy distribution with the absolute bare minimum settings (no custom headers, default SSL) to see if that flies through. If it does, the problem is isolated to your specific config. If it *also* hangs, it's almost certainly an AWS backend issue in your region. Let us know what you find!
If it's not measurable, it's not marketing.