4 comments

  • taneliv48 minutes ago
    While Jepsen (and this article) is focused on behavior under node failure and network partitions, this caught my eye:<p>&gt; It also exhibits Stale Read, Lost Update, and other forms of G-single in healthy clusters<p>This looks like quite a fundamental issue.
  • linsomniac2 hours ago
    I really like glaera for low volume clustering, because of the true multi-master nature. I&#x27;ve been using it for over a decade on a clustered mail server for storing account information, and more recently I&#x27;ve pumped the log information in there so each user can see their related log messages, for a user base of around 6,000 users, and it&#x27;s been a real workhorse.
  • constructrurl2 hours ago
    [dead]
  • linsomniac2 hours ago
    I realize that we like to use the page title here on HN, but this really should be something like &quot;Data loss cases with MariaDB Glaera Cluster 12.1.2&quot;.
    • hu31 hour ago
      One of the reasons is that this kind of title editorialization fosters generic commentaries in reaction to titles.