c/mirror_zfs
1
0
Fork 0
mirror of https://git.proxmox.com/git/mirror_zfs.git synced 2026-05-24 11:18:52 +03:00
Code Issues Packages Projects Releases Wiki Activity

Teach zpool scrub to scrub only blocks in error log

Browse Source 
Added a flag '-e' in zpool scrub to scrub only blocks in error log. A
user can pause, resume and cancel the error scrub by passing additional
command line arguments -p -s just like a regular scrub. This involves
adding a new flag, creating new libzfs interfaces, a new ioctl, and the
actual iteration and read-issuing logic. Error scrubbing is executed in
multiple txg to make sure pool performance is not affected.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Co-authored-by: TulsiJain tulsi.jain@delphix.com
Signed-off-by: George Amanakis <gamanakis@gmail.com>
Closes #8995
Closes #12355
This commit is contained in:
George Amanakis
2021-12-17 21:35:28 +01:00
committed by Brian Behlendorf
parent e34e15ed6d
commit 482eeef804
29 changed files with 1602 additions and 71 deletions
Download Patch File Download Diff File Expand all files Collapse all files
include/sys/dsl_scan.h
+26 -1
View File
@@ -29,6 +29,7 @@
#include <sys/zfs_context.h>
#include <sys/zio.h>
#include <sys/zap.h>
#include <sys/ddt.h>
#include <sys/bplist.h>
@@ -78,6 +79,21 @@ typedef enum dsl_scan_flags {
#define DSL_SCAN_FLAGS_MASK (DSF_VISIT_DS_AGAIN)
typedef struct dsl_errorscrub_phys {
uint64_t dep_func; /* pool_scan_func_t */
uint64_t dep_state; /* dsl_scan_state_t */
uint64_t dep_cursor; /* serialized zap cursor for tracing progress */
uint64_t dep_start_time; /* error scrub start time, unix timestamp */
uint64_t dep_end_time; /* error scrub end time, unix timestamp */
uint64_t dep_to_examine; /* total error blocks to be scrubbed */
uint64_t dep_examined; /* blocks scrubbed so far */
uint64_t dep_errors; /* error scrub I/O error count */
uint64_t dep_paused_flags; /* flag for paused */
} dsl_errorscrub_phys_t;
#define ERRORSCRUB_PHYS_NUMINTS (sizeof (dsl_errorscrub_phys_t) \
/ sizeof (uint64_t))
/*
* Every pool will have one dsl_scan_t and this structure will contain
* in-memory information about the scan and a pointer to the on-disk
@@ -151,11 +167,15 @@ typedef struct dsl_scan {
uint64_t scn_avg_zio_size_this_txg;
uint64_t scn_zios_this_txg;
/* zap cursor for tracing error scrub progress */
zap_cursor_t errorscrub_cursor;
/* members needed for syncing scan status to disk */
dsl_scan_phys_t scn_phys; /* on disk representation of scan */
dsl_scan_phys_t scn_phys_cached;
avl_tree_t scn_queue; /* queue of datasets to scan */
uint64_t scn_queues_pending; /* outstanding data to issue */
/* members needed for syncing error scrub status to disk */
dsl_errorscrub_phys_t errorscrub_phys;
} dsl_scan_t;
typedef struct dsl_scan_io_queue dsl_scan_io_queue_t;
@@ -171,8 +191,12 @@ int dsl_scan_cancel(struct dsl_pool *);
int dsl_scan(struct dsl_pool *, pool_scan_func_t);
void dsl_scan_assess_vdev(struct dsl_pool *dp, vdev_t *vd);
boolean_t dsl_scan_scrubbing(const struct dsl_pool *dp);
int dsl_scrub_set_pause_resume(const struct dsl_pool *dp, pool_scrub_cmd_t cmd);
boolean_t dsl_errorscrubbing(const struct dsl_pool *dp);
boolean_t dsl_errorscrub_active(dsl_scan_t *scn);
void dsl_scan_restart_resilver(struct dsl_pool *, uint64_t txg);
int dsl_scrub_set_pause_resume(const struct dsl_pool *dp,
pool_scrub_cmd_t cmd);
void dsl_errorscrub_sync(struct dsl_pool *, dmu_tx_t *);
boolean_t dsl_scan_resilvering(struct dsl_pool *dp);
boolean_t dsl_scan_resilver_scheduled(struct dsl_pool *dp);
boolean_t dsl_dataset_unstable(struct dsl_dataset *ds);
@@ -184,6 +208,7 @@ void dsl_scan_ds_clone_swapped(struct dsl_dataset *ds1, struct dsl_dataset *ds2,
struct dmu_tx *tx);
boolean_t dsl_scan_active(dsl_scan_t *scn);
boolean_t dsl_scan_is_paused_scrub(const dsl_scan_t *scn);
boolean_t dsl_errorscrub_is_paused(const dsl_scan_t *scn);
void dsl_scan_freed(spa_t *spa, const blkptr_t *bp);
void dsl_scan_io_queue_destroy(dsl_scan_io_queue_t *queue);
void dsl_scan_io_queue_vdev_xfer(vdev_t *svd, vdev_t *tvd);