You are viewing a single comment's thread from:

RE: Imagehoster cache purge for Cloudflare

in HiveDevs4 years ago

Losing the images is not great as happened recently. Hope you can find solutions that work for us. Cooperation between the dapp teams is vital to get the best results.

Sort:  

Could you elaborate or remind, where/when losing images happened or circumstances around that incident?

Last time was last Friday morning (in UK). Profile images and most of those in posts were not visible for at least a couple of hours in peakd and hive.blog at least. I saw people reporting that they could not upload images on various discord channels. This has happened before. I assume that eventually someone was available to fix it and it was okay later on. That is all I know.

The images are not gone. There simply don't load on the front ends because of the CDN.

I assumed they were preserved as they came back, but not having them visible confuses a lot of people. I realise that making the system resilient is hard. We saw a lot of big sites taken out totally by a CDN issue recently. Hive dapps have limited resources to pay for expensive services, but I know we have some clever people working on solution.

Ah, it was only on those frontends, ecency.com didn't have any problems since it is using other instance.