[solved] Openpaper.work is down (datacenter fire)

Hello,

It appears that the datacenter where openpaper.work is hosted was on fire this morning (OVH Strasbourg): https://www.datacenterknowledge.com/uptime/ceo-says-fire-has-destroyed-ovh-s-strasbourg-data-center-sbg2

Both VPS and object storage for openpaper.work are in SBG5. Since there are no news specifically about SGB5, I can only assume everything is fine and will be running again as soon as they turn the datacenter back on.

Broken:

  • openpaper.work: (back online: 2021/3/22)
    • the scanner database
    • usage statistics of Paperwork
    • automated bug reports
  • Paperwork for Windows download
  • doc.openpaper.work
  • download.openpaper.work
  • build pipelines
  • my emails …

Working:

  • Paperwork for GNU/Linux download (Flatpak and Pypi)
  • forum.openpaper.work (obviously :wink: ) (except for email notifications)
  • translations.openpaper.work (except for email notifications)
  • builder.openpaper.work (Flatpak repository)
  • Gitlab

To be honest, I’m unsure where my backups are hosted. I fear they are on SBG5 too.
The worst case scenario here is that we loose the scanner database, Paperwork usage statistics, and the object storage (all of the Windows builds, documentation builds and data files built over time). I have an old copy of the scanner database on my own computer (a few months old).

Best regards,

Emails notifications for forum.openpaper.work are back online.

OVH has been able to restore the 2 VPS managing openpaper.work and my emails. No data lost here \o/.

Still waiting to see if we get the object storages back. They are required for:

  • build data & download data
  • the build pipelines
  • the build results (Windows binaries, documentation, etc)
  • doc.openpaper.work

(nothing that can’t be replaced)

Object storages are OK. From what I can tell, no data has been lost.

Now that I have fiber optic at my home, to spare money, I’m going to move all the services to my personal server. So they may be unavailable from time to time for short amount of time.