Skip to contentSkip to conversation

btrfs: Allow read-only mount with corrupted extent tree

PATCH In review Dāvis Mosāns 2 participants 3 messages Aug 11, 2021 – Aug 12, 2021 Btrfs file system

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 btrfs: Allow read-only mount with corrupted extent tree

Currently if there's any corruption at all in extent tree (eg. even single bit) then mounting will fail with:
"failed to read block groups: -5" (-EIO)
It happens because we immediately abort on first error when searching in extent tree for block groups.

Now with this patch if ignorebadroots option is specified then we handle such case and continue by removing already created block groups and creating dummy block groups.

---
 fs/btrfs/block-group.c | 20 ++++++++++++++++++++
 fs/btrfs/disk-io.c     |  4 ++--
 fs/btrfs/disk-io.h     |  2 ++
 3 files changed, 24 insertions(+), 2 deletions(-)
fs/btrfs/block-group.c +20 -0
index 9e7d9d0c763d..80b9bb9afb8c 100644
@@ -2212,6 +2212,26 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info)
 	ret = check_chunk_block_group_mappings(info);
 error:
 	btrfs_free_path(path);
+
+	if (ret == -EIO && btrfs_test_opt(info, IGNOREBADROOTS)) {
+
+		if (btrfs_super_log_root(info->super_copy) != 0) {
+			btrfs_warn(info, "Ignoring tree-log replay due to extent tree corruption!");
+			btrfs_set_super_log_root(info->super_copy, 0);
+		}
+
+		btrfs_put_block_group_cache(info);
+		btrfs_stop_all_workers(info);
+		btrfs_free_block_groups(info);
+		ret = btrfs_init_workqueues(info, NULL);
+		if (ret)
+			return ret;
+		ret = btrfs_init_space_info(info);
+		if (ret)
+			return ret;
+		return fill_dummy_bgs(info);
+	}
+
 	return ret;
 }
 
fs/btrfs/disk-io.c +2 -2
index a59ab7b9aea0..b1ad9c85d578 100644
@@ -2115,7 +2115,7 @@ static int read_backup_root(struct btrfs_fs_info *fs_info, u8 priority)
 }
 
 /* helper to cleanup workers */
-static void btrfs_stop_all_workers(struct btrfs_fs_info *fs_info)
+void btrfs_stop_all_workers(struct btrfs_fs_info *fs_info)
 {
 	btrfs_destroy_workqueue(fs_info->fixup_workers);
 	btrfs_destroy_workqueue(fs_info->delalloc_workers);
@@ -2283,7 +2283,7 @@ static void btrfs_init_qgroup(struct btrfs_fs_info *fs_info)
 	mutex_init(&fs_info->qgroup_rescan_lock);
 }
 
-static int btrfs_init_workqueues(struct btrfs_fs_info *fs_info,
+int btrfs_init_workqueues(struct btrfs_fs_info *fs_info,
 		struct btrfs_fs_devices *fs_devices)
 {
 	u32 max_active = fs_info->thread_pool_size;
fs/btrfs/disk-io.h +2 -0
index 0e7e9526b6a8..41348c8d3f9a 100644
@@ -139,6 +139,8 @@ int btrfs_get_free_objectid(struct btrfs_root *root, u64 *objectid);
 int btrfs_init_root_free_objectid(struct btrfs_root *root);
 int __init btrfs_end_io_wq_init(void);
 void __cold btrfs_end_io_wq_exit(void);
+void btrfs_stop_all_workers(struct btrfs_fs_info *fs_info);
+int btrfs_init_workqueues(struct btrfs_fs_info *fs_info, struct btrfs_fs_devices *fs_devices);
 
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
 void btrfs_set_buffer_lockdep_class(u64 objectid,
Signature · 1 line
2.32.0
#1 lore raw
Quoted 10 lines · Dāvis Mosāns

Currently if there's any corruption at all in extent tree (eg. even single bit) then mounting will fail with:
"failed to read block groups: -5" (-EIO)
It happens because we immediately abort on first error when searching in extent tree for block groups.

Now with this patch if ignorebadroots option is specified then we handle such case and continue by removing already created block groups and creating dummy block groups.

Already done and queue'ed up for the next release

btrfs: rescue: allow ibadroots to skip bad extent tree when reading block group items

thanks,

Josef

#2 lore raw

trešd., 2021. g. 11. aug., plkst. 23:41 — lietotājs Josef Bacik (<[email protected]>) rakstīja:

Quoted 13 lines · Dāvis Mosāns
Quoted 10 lines

Currently if there's any corruption at all in extent tree (eg. even single bit) then mounting will fail with:
"failed to read block groups: -5" (-EIO)
It happens because we immediately abort on first error when searching in extent tree for block groups.

Now with this patch if ignorebadroots option is specified then we handle such case and continue by removing already created block groups and creating dummy block groups.

Quoted 4 lines

Already done and queue'ed up for the next release

btrfs: rescue: allow ibadroots to skip bad extent tree when reading block group items

Nice!
Originally I submitted this patch in March, but it went nowhere so resent. Anyway I tested that other patch and it has limitation that it won't handle log replay error so need to use both rescue=ignorebadroots,nologreplay
But other than it seems to work fine.

Here's output when not using nologreplay

[ 1886.281376] BTRFS info (device sde): ignoring bad roots
[ 1886.281381] BTRFS info (device sde): disk space caching is enabled
[ 1886.281383] BTRFS info (device sde): has skinny extents
[ 1886.365608] BTRFS info (device sde): bdev /dev/sdq errs: wr 0, rd
0, flush 0, corrupt 473, gen 0
[ 1886.365618] BTRFS info (device sde): bdev /dev/sdi errs: wr 31626,
rd 18765, flush 178, corrupt 5841, gen 0
[ 1886.365623] BTRFS info (device sde): bdev /dev/sdo errs: wr 6867,
rd 2640, flush 178, corrupt 1066, gen 0
[ 1900.249267] BTRFS warning (device sde): checksum verify failed on
21057125941248 wanted 0x5a4526a7 found 0x25949991 level 0
[ 1902.323362] BTRFS error (device sde): parent transid verify failed
on 21057111523328 wanted 2262739 found 2262698
[ 1902.519338] BTRFS error (device sde): bad tree block start, want
21057108836352 have 524288
[ 1902.940022] BTRFS warning (device sde): checksum verify failed on
21057097302016 wanted 0x8b2501e9 found 0x5b8ab9a3 level 0
[ 1902.946715] BTRFS error (device sde): parent transid verify failed
on 21057097302016 wanted 2262739 found 2262696
[ 1902.950727] BTRFS info (device sde): start tree-log replay
[ 1911.289738] BTRFS warning (device sde): checksum verify failed on
21057127661568 wanted 0xd1506ed9 found 0x22ab750a level 0
[ 1911.293266] BTRFS warning (device sde): checksum verify failed on
21057107017728 wanted 0x9120eaee found 0x35b2df28 level 0
[ 1911.326093] BTRFS: error (device sde) in __btrfs_free_extent:3069:
errno=-5 IO failure
[ 1911.326101] BTRFS: error (device sde) in
btrfs_run_delayed_refs:2150: errno=-5 IO failure
[ 1911.326106] BTRFS warning (device sde): Skipping commit of aborted
transaction.
[ 1911.326108] BTRFS: error (device sde) in cleanup_transaction:1945:
errno=-5 IO failure
[ 1911.326190] BTRFS warning (device sde): checksum verify failed on
21057107017728 wanted 0x9120eaee found 0xd00e16cf level 0
[ 1911.326317] BTRFS warning (device sde): checksum verify failed on
21057107017728 wanted 0x9120eaee found 0xd00e16cf level 0
[ 1911.326922] BTRFS: error (device sde) in btrfs_replay_log:2417:
errno=-5 IO failure (Failed to recover log tree)
[ 1912.160913] BTRFS error (device sde): parent transid verify failed
on 21057111736320 wanted 2262739 found 2262698
[ 1914.666177] BTRFS error (device sde): open_ctree failed

Best regards,
Dāvis