Over the last couple of days I’ve been doing some behind-the-scenes maintenance on ClarionHub to fix a few issues that had built up over time.
The main problem was that our internal Discourse backups had stopped running. This appears to be related to the original S3 account that Brahn set up years ago — that account is no longer active, so anything relying on it had quietly failed in the background.
To get everything working again, I’ve moved all backups (and now uploads as well) onto a new S3 bucket under my own AWS account. The Hub is now successfully backing up again, and new uploads are being stored correctly.
There is a chance that some older posts may have missing images. A portion of the original upload files were stored on the old S3 account, and those are no longer reachable. I’m waiting for AWS to finish generating an inventory so I can run a scan and work out exactly how many files are affected. Early estimates suggest it might be a few thousand images, but I’ll confirm once the analysis is complete.
I’ll also be looking at ways to minimise the ongoing S3 storage costs, likely via some Cloudflare optimisation.
For now:
– Backups are working again
– Uploads are fully functional
– More info on any missing legacy images will follow
Thanks for your patience while I got this sorted! If you notice anything odd on the site, feel free to drop me a message.