downtime tomorrow at 1200 UTC


once again, I have to announce a downtime of bilbo, our server for mail, mailing lists and downloads. As the recently exchanged SSD died again after three (!) days, weed need to investigate further. The SSD killing rate of the machine is beyond any imagination, and althouhg the migration to a completely new system is in the works already, we'd like to have some stability until that is dealt with.

Therefore, our ISP will change the port, cable and disk tomorrow,

  Thursday, at 1200 UTC

which will lead to a downtime of the mentioned services, of approximately 30-60 minutes.

I am sorry for those annoying downtimes and inconveniences. The new infra concept foresees splitting up services on various machines, so one system down does not affect large parts of our system. Blocker is - as stated in my previous e-mail - a severe lack of documentation so far I'm tryin to compensate now, which delays things quite a lot.

As usual, the infra team is available at #tdf-infra for questions during the downtime, or at via e-mail before.