c/mirror_zfs
1
0
Fork 0
mirror of https://git.proxmox.com/git/mirror_zfs.git synced 2026-05-22 18:40:43 +03:00
Code Issues Packages Projects Releases Wiki Activity

Implement a new type of zfs receive: corrective receive (-c)

Browse Source 
This type of recv is used to heal corrupted data when a replica
of the data already exists (in the form of a send file for example).
With the provided send stream, corrective receive will read from
disk blocks described by the WRITE records. When any of the reads
come back with ECKSUM we use the data from the corresponding WRITE
record to rewrite the corrupted block.

Reviewed-by: Paul Dagnelie <pcd@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Paul Zuchowski <pzuchowski@datto.com>
Signed-off-by: Alek Pinchuk <apinchuk@axcient.com>
Closes #9372
This commit is contained in:
Alek P
2022-07-28 18:52:46 -04:00
committed by GitHub
parent 5fae33e047
commit e8cf3a4f76
28 changed files with 1233 additions and 74 deletions
Download Patch File Download Diff File Expand all files Collapse all files
module/zfs/spa_errlog.c
+110 -4
View File
@@ -22,6 +22,7 @@
* Copyright (c) 2006, 2010, Oracle and/or its affiliates. All rights reserved.
* Copyright (c) 2013, 2014, Delphix. All rights reserved.
* Copyright (c) 2021, George Amanakis. All rights reserved.
* Copyright (c) 2019 Datto Inc.
*/
/*
@@ -68,11 +69,13 @@
#include <sys/dmu_objset.h>
#include <sys/dbuf.h>
#define NAME_MAX_LEN 64
/*
* spa_upgrade_errlog_limit : A zfs module parameter that controls the number
* of on-disk error log entries that will be converted to the new
* format when enabling head_errlog. Defaults to 0 which converts
* all log entries.
* of on-disk error log entries that will be converted to the new
* format when enabling head_errlog. Defaults to 0 which converts
* all log entries.
*/
static uint32_t spa_upgrade_errlog_limit = 0;
@@ -511,6 +514,103 @@ get_errlist_size(spa_t *spa, avl_tree_t *tree)
}
#endif
/*
* If a healed bookmark matches an entry in the error log we stash it in a tree
* so that we can later remove the related log entries in sync context.
*/
static void
spa_add_healed_error(spa_t *spa, uint64_t obj, zbookmark_phys_t *healed_zb)
{
char name[NAME_MAX_LEN];
if (obj == 0)
return;
bookmark_to_name(healed_zb, name, sizeof (name));
mutex_enter(&spa->spa_errlog_lock);
if (zap_contains(spa->spa_meta_objset, obj, name) == 0) {
/*
* Found an error matching healed zb, add zb to our
* tree of healed errors
*/
avl_tree_t *tree = &spa->spa_errlist_healed;
spa_error_entry_t search;
spa_error_entry_t *new;
avl_index_t where;
search.se_bookmark = *healed_zb;
mutex_enter(&spa->spa_errlist_lock);
if (avl_find(tree, &search, &where) != NULL) {
mutex_exit(&spa->spa_errlist_lock);
mutex_exit(&spa->spa_errlog_lock);
return;
}
new = kmem_zalloc(sizeof (spa_error_entry_t), KM_SLEEP);
new->se_bookmark = *healed_zb;
avl_insert(tree, new, where);
mutex_exit(&spa->spa_errlist_lock);
}
mutex_exit(&spa->spa_errlog_lock);
}
/*
* If this error exists in the given tree remove it.
*/
static void
remove_error_from_list(spa_t *spa, avl_tree_t *t, const zbookmark_phys_t *zb)
{
spa_error_entry_t search, *found;
avl_index_t where;
mutex_enter(&spa->spa_errlist_lock);
search.se_bookmark = *zb;
if ((found = avl_find(t, &search, &where)) != NULL) {
avl_remove(t, found);
kmem_free(found, sizeof (spa_error_entry_t));
}
mutex_exit(&spa->spa_errlist_lock);
}
/*
* Removes all of the recv healed errors from both on-disk error logs
*/
static void
spa_remove_healed_errors(spa_t *spa, avl_tree_t *s, avl_tree_t *l, dmu_tx_t *tx)
{
char name[NAME_MAX_LEN];
spa_error_entry_t *se;
void *cookie = NULL;
ASSERT(MUTEX_HELD(&spa->spa_errlog_lock));
while ((se = avl_destroy_nodes(&spa->spa_errlist_healed,
&cookie)) != NULL) {
remove_error_from_list(spa, s, &se->se_bookmark);
remove_error_from_list(spa, l, &se->se_bookmark);
bookmark_to_name(&se->se_bookmark, name, sizeof (name));
kmem_free(se, sizeof (spa_error_entry_t));
(void) zap_remove(spa->spa_meta_objset,
spa->spa_errlog_last, name, tx);
(void) zap_remove(spa->spa_meta_objset,
spa->spa_errlog_scrub, name, tx);
}
}
/*
* Stash away healed bookmarks to remove them from the on-disk error logs
* later in spa_remove_healed_errors().
*/
void
spa_remove_error(spa_t *spa, zbookmark_phys_t *zb)
{
char name[NAME_MAX_LEN];
bookmark_to_name(zb, name, sizeof (name));
spa_add_healed_error(spa, spa->spa_errlog_last, zb);
spa_add_healed_error(spa, spa->spa_errlog_scrub, zb);
}
/*
* Return the number of errors currently in the error log. This is actually the
* sum of both the last log and the current log, since we don't know the union
@@ -887,7 +987,7 @@ void
sync_error_list(spa_t *spa, avl_tree_t *t, uint64_t *obj, dmu_tx_t *tx)
{
spa_error_entry_t *se;
char buf[64];
char buf[NAME_MAX_LEN];
void *cookie;
if (avl_numnodes(t) == 0)
@@ -992,6 +1092,7 @@ spa_errlog_sync(spa_t *spa, uint64_t txg)
*/
if (avl_numnodes(&spa->spa_errlist_scrub) == 0 &&
avl_numnodes(&spa->spa_errlist_last) == 0 &&
avl_numnodes(&spa->spa_errlist_healed) == 0 &&
!spa->spa_scrub_finished) {
mutex_exit(&spa->spa_errlist_lock);
return;
@@ -1006,6 +1107,11 @@ spa_errlog_sync(spa_t *spa, uint64_t txg)
tx = dmu_tx_create_assigned(spa->spa_dsl_pool, txg);
/*
* Remove healed errors from errors.
*/
spa_remove_healed_errors(spa, &last, &scrub, tx);
/*
* Sync out the current list of errors.
*/