How is data stored on Inaturalist?

[ahospers]

Probably Azure

@kueda

Update: a disk that we share between several severs stopped working this morning for reasons that are still a mystery. We mostly use that disk for caching files, i.e. non-essential ephemera, but there are a few kinds of data from older parts of the site that were hanging around in there. Patrick got a new disk working and we’re talking with the folks at Azure to see if they can explain what happened to the old disk and whether we can recover the data, but there’s a good chance some of it was lost.

Here’s the more permanent data that may have been lost:

1. Project Assets: this is a feature of trusted traditional projects that allowed them to upload arbitrary files like banner images and KML map overlays. Very few projects will be affected by this, but unfortunately the Vermont Atlas of Life is one of them, so FYI, @kpmcfarland.
2. Taxon Ranges: these actually probably won’t be affected b/c the data in an attached file gets loaded into the database, so loss of the original file doesn’t represent a loss of data

Again, we’re working on restoring this stuff, but at this point we’re waiting on Azure. We’re sorry if this has affected you, and if you need help getting projects set up again, please let us know (though it will probably be easiest for all concerned to just not use these customization features). The more ephemeral types of data that were on that disk include old observation exports, observation imports and exports that were in progress when the disk died, and DarwinCore archives.

https://forum.inaturalist.org/t/how-confident-can-we-be-that-inaturalist-data-will-be-preserved/3248/13?u=ahospers