diff options
| author | Alexander Motin <alexander.motin@TrueNAS.com> | 2025-12-18 18:21:45 +0000 |
|---|---|---|
| committer | Tony Hutter <hutter2@llnl.gov> | 2026-02-05 21:48:31 +0000 |
| commit | 35ee242abcba40afc7d91e7c9f8da9c3fa29e646 (patch) | |
| tree | 5f7c5ca0e829551af84aebc42791307fe0fd65c7 /module | |
| parent | ccb7c82aa15b1b3cb5b9e24e1ce56ec5aadbcd0d (diff) | |
Reduce minimal scrub/resilver times
With higher throughput and lower latency of modern devices ZFS can
happily live with pretty short (fractions of a second) TXGs. But
the two decade old multi-second minimal time limits can almost stop
payload writes by extending TXGs beyond dirty data limits of ARC
ability to amortize it.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Alexander Motin <alexander.motin@TrueNAS.com>
Closes #18060
Diffstat (limited to 'module')
| -rw-r--r-- | module/zfs/dsl_scan.c | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/module/zfs/dsl_scan.c b/module/zfs/dsl_scan.c index 4a9e831dc434..98bdf477162e 100644 --- a/module/zfs/dsl_scan.c +++ b/module/zfs/dsl_scan.c @@ -189,7 +189,7 @@ static uint_t zfs_scan_mem_lim_fact = 20; static uint_t zfs_scan_mem_lim_soft_fact = 20; /* minimum milliseconds to scrub per txg */ -static uint_t zfs_scrub_min_time_ms = 1000; +static uint_t zfs_scrub_min_time_ms = 750; /* minimum milliseconds to obsolete per txg */ static uint_t zfs_obsolete_min_time_ms = 500; @@ -198,7 +198,7 @@ static uint_t zfs_obsolete_min_time_ms = 500; static uint_t zfs_free_min_time_ms = 500; /* minimum milliseconds to resilver per txg */ -static uint_t zfs_resilver_min_time_ms = 3000; +static uint_t zfs_resilver_min_time_ms = 1500; static uint_t zfs_scan_checkpoint_intval = 7200; /* in seconds */ int zfs_scan_suspend_progress = 0; /* set to prevent scans from progressing */ |