RAIDZ_TEST(1)
β¨ RAIDZ_TEST(1) β¨
okay bestie so this is a page from the General Commands Manual, and it's all about raidz_test π
πΈ NAME πΈ
raidz_test β the raidz implementation verification and benchmarking tool π§
π¦ SYNOPSIS π¦
raidz_test [-StBevTD] [-a ashift] [-o zio_off_shift]
[-d raidz_data_disks] [-s zio_size_shift] [-r reflow_offset]
π DESCRIPTION π
okay so the whole point of this tool is to run every supported raidz implementation and then verify that all the methods actually produce the right results. no cap. π«Ά
it also has a parameter sweep option where every single parameter that affects a RAID-Z block gets verified β think ashift size, data offset, data size, all of it, etc. etc. the tool also supports a benchmarking mode and you get that with the -B option β¨
π OPTION π
-hβ print a help summary. short, sweet, iconic π-aashift (default: 9) β the ashift value. that's it, that's the whole vibe.-ozio_off_shift (default: 0) β ZIO offset for each raidz block. the offset's actual value is2^zio_off_shiftπ-draidz_data_disks (default: 8) β how many raidz data disks to use. any extra disks you throw in get used for parity. periodt.-szio_size_shift (default: 19) β the size of data for a raidz block. the real size is2^zio_size_shiftβ-rreflow_offset (default: uint max) β sets the raidz expansion offset. the expanded raidz map allocation function will produce different map configurations depending on this value π-S(weep) β sweep the parameter space while verifying the raidz implementations. this one will exhaust all most of the valid values for the-aodsoptions. heads up: runtime using this option will be LONG. like, go make a snack long π-t(imeout) β wall time for the sweep test in seconds. and ngl the actual runtime could be longer than what you asked for.-B(enchmark) β all implementations get benchmarked using increasing per disk data size. results come out as throughput per disk, measured in MiB/s π₯-e(xpansion) β use the expanded raidz map allocation function.-v(erbose) β increase verbosity. say more, bestie π£οΈ-T(est the test) β debugging option: fail all the tests. this exists to check whether the tests would properly verify bit-exactness. it's giving "chaos but make it scientific" π-D(ebug) β debugging option: attachgdb(1)whenSIGSEGVorSIGABRTare received. so when it crashes, it crashes fancy π
fr fr,-Sand-tare not a quick little errand. plan your day around them.
π SEE ALSO π
and that's the whole page, bestie π this was brought to you by OpenZFS, dated May 26, 2021 β love that for us, mwah πβ¨π
π¦ where this page comes from
- alpine 3.22: zfs-doc 2.3.9-r0 Β· CDDL-1.0
- alpine 3.23: zfs-doc 2.4.4-r0 Β· CDDL-1.0
- alpine 3.24: zfs-doc 2.4.4-r0 Β· CDDL-1.0
- debian trixie: zfs-test 2.3.9-0+deb13u1 Β· CDDL-1.0
- nixos 26.05: zfs-user 2.4.4
- nixpkgs unstable: zfs-user 2.4.4
- ubuntu questing: zfs-test 2.3.4-1ubuntu2
- ubuntu resolute: zfs-test 2.4.1-1ubuntu5 Β· CDDL-1.0
- ubuntu resolute: zfs-test 2.4.1-1ubuntu5.1 Β· CDDL-1.0
- ubuntu stonking: zfs-test 2.4.4-1ubuntu4 Β· CDDL-1.0
- π the page's license: CDDL-1.0
- Copyright (c) 2016 Gvozden NeΕ‘koviΔ. All rights reserved
- π girlified by deepseek-ai/DeepSeek-V4.1-Flash on : a language model's rewrite of the original, under its terms