You are reading a single comment by @mespilus and its replies. Click here to read the full conversation.
  • The outage this morning was a subtle change.

    From 09:13 - 09:45 BST I created and moved a couple of servers.

    I've nuked the Ubuntu 14.04 API and database servers, including the PostgreSQL 9.2 instance that had run for the last 7 years with no upgrades.

    I've created a new and hardened Ubuntu 20.04 server with PostgreSQL 14 and the latest version of microcosm.

    The new server was needed as we were facing a disk space crunch on the old server and also running way too hot (hence not being as snappy as the site used to be).

    So we've also gone from 4 virtual CPU cores to 8 dedicated CPU cores.

    During this move I did lose 2 comments due to a timing error on starting the database backup and the existing service not being shut down quickly enough. Apologies to whomever made those... they're gone in the very quick shuffle of the scripts working and I only know they happened as the nginx logs in front of everything showed 2 succcessful HTTP POST requests.

    Anyhow, server surgery done.

  • Thanks for the explanation.
    When I receive the server error notification, I always expect it is something 'wrong' at my end,
    be it laptop or mobile,
    but,
    then remembered reading one of your recent posts about needing to upgrade stuff,
    (beyond my comprehension).
    Thanks for the update, and all the work you do behind the scenes to keep this forum going.
    (If you ever get to a Wests there are some cracking Newton Wonder apples on offer at the moment.
    I'm certain you could make yourself a passable apple crumble in your new kitchen).

About

Avatar for mespilus @mespilus started