xfs_scrub_all(8)

🎀 xfs_scrub_all(8) 🎀 — NAME ✨

okay so the manual name is xfs_scrub_all, and its whole job is to scrub all mounted XFS filesystems 💅

🌸 NAME 🌸

xfs_scrub_all - scrub all mounted XFS filesystems ✨

💖 SYNOPSIS 💖

xfs_scrub_all [ -hxV ]

🦋 DESCRIPTION 🦋

okay bestie, so xfs_scrub_all tries to read and check all the metadata on every mounted XFS filesystem 🫶 the online scrub gets done by the xfs_scrub tool, and you can either run it directly or let systemd kick it off in a restricted way 🔥

here's the lowkey clever part 🧚 every mounted filesystem gets mapped to its physical storage device, so scrubs can run in parallel as long as no two scrubbers are touching the same device at the same time. no cap, that's giving efficiency 💜

👑 OPTIONS 👑

• --auto-media-scan-interval — automatically turns on the file data scan (aka the -x flag) if it hasn't run within the interval you give it 🌈 the interval can be a floating point number with an optional unit suffix. the units you can use are y, q, mo, w, d, h, m, and s, which mean years, 90-day quarters, 30-day months, weeks, days, hours, minutes, and seconds, in that order 🍓 if you don't specify a unit, it defaults to seconds
• --auto-media-scan-stamp — the path to a file that records the last time the media scan ran 📝 it defaults to /var/lib/xfsprogs/xfs_scrub_all_media.stamp
• -h — displays help 💁‍♀️
• -x — reads all file data extents to look for disk errors 🔍
• -V — prints the version number and exits 🫧

🌈 EXIT CODE 🌈

okay so the exit code xfs_scrub_all returns is the sum of the following conditions, so you can decode what went wrong by adding up the numbers 🧮

• 0 — no errors 🥰
• 4 — file system errors left uncorrected 😭
• 8 — operational error 💀
• 16 — usage or syntax error 🙈

these are the same error codes that xfs_scrub returns too ✨

🧚 SEE ALSO 🧚

xfs_scrub(8).

bestie, that's the whole page 🫶 this was the xfsprogs manual, from the xfs_scrub_all(8) page of the System Manager's Manual 💖🦋 slay forever 💅✨