Grown-up ZFS Data Corruption Bug

Hey hey! We recently tripped over a ZFS data corruption bug–introduced over 18 years ago! Bryan and Adam discuss with members of the Oxide team as well as Matt Ahrens, the co-inventor of ZFS.
In addition to Bryan Cantrill (bsky.app/profile/bcantrill.bsky.social) and Adam Leventhal (bsky.app/profile/ahl.bsky.social), speakers included Alan Hanson, Matt Keeter (hachyderm.io/@mjk), Andy Fiddaman, James MacMahon, and special guest, Matt Ahrens.
Previously, on Oxide and Friends:
• OxF s4e6 - Crucible: the Oxide Storage Service (oxide-and-friends.transistor.fm/episodes/cruci…)
• OxF s5e28 - Systems Software in the Large (oxide-and-friends.transistor.fm/episodes/syste…)
Some of the topics we hit on, in the order that we hit them:
• ZFS fsync can trigger ZIL transaction reordering and data corruption (illumos.org/issues/17734)
• RFD 177: Implementation of Data Storage (rfd.shared.oxide.computer/rfd/0177)
• the "fix" that introduced data corruption (illumos.org/opensolaris/bugdb/bug.html#!6535160)
• PRs needed!
If we got something wrong or missed something, please file a PR! Our next show will likely be on Monday at 5p Pacific Time on our Discord server; stay tuned to our Mastodon feeds for details, or subscribe to this calendar (calendar.google.com/calendar/ical/c_318925f418…). We'd love to have you join us, as we always love to hear from new speakers!