btrfs: rescue: allow ibadroots to skip bad extent tree when reading block group items
PATCH v2 1/1 btrfs: rescue: allow ibadroots to skip bad extent tree when reading block group items
When extent tree gets corrupted, normally it's not extent tree root, but one toasted tree leaf/node.
In that case, rescue=ibadroots mount option won't help as it can only handle the extent tree root corruption.
This patch will enhance the behavior by:
- Allow fill_dummy_bgs() to ignore -EEXIST error
This means we may have some block group items read from disk, but then hit some error halfway.
- Fallback to fill_dummy_bgs() if any error gets hit in btrfs_read_block_groups()
Of course, this still needs rescue=ibadroots mount option.
With that, rescue=ibadroots can handle extent tree corruption more gracefully and allow a better recover chance.
---
Changelog:
v2:
- Don't try to fill with dummy block groups when we hit ENOMEM
--- fs/btrfs/block-group.c | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-)
fs/btrfs/block-group.c +13 -1
@@ -2105,11 +2105,16 @@ static int fill_dummy_bgs(struct btrfs_fs_info *fs_info) bg->used = em->len; bg->flags = map->type; ret =btrfs_add_block_group_cache (fs_info, bg); -if (ret) { +/* +* We may have some block groups filled already, thus ignore +* the -EEXIST error. +*/ +if (ret && ret != -EEXIST ) {btrfs_remove_free_space_cache (bg);btrfs_put_block_group (bg);break ; } + ret =0 ;btrfs_update_space_info (fs_info, bg->flags, em->len, em->len,0 ,0 , &space_info); bg->space_info = space_info;
@@ -2212,6 +2217,13 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info) ret =check_chunk_block_group_mappings (info); error:btrfs_free_path (path); +/* +* Either we have no extent_root (already implies IGNOREBADROOTS), or +* we hit error and have IGNOREBADROOTS, then we can try to use dummy +* bgs. +*/ +if (!info->extent_root || (ret &&btrfs_test_opt (info,IGNOREBADROOTS ))) + ret =fill_dummy_bgs (info);return ret; }
Signature · 1 line
2.32.0
Quoted 58 lines · Qu Wenruo
When extent tree gets corrupted, normally it's not extent tree root, but one toasted tree leaf/node.
In that case, rescue=ibadroots mount option won't help as it can only handle the extent tree root corruption.
This patch will enhance the behavior by:
- Allow
fill_dummy_bgs()to ignore-EEXISTerrorThis means we may have some block group items read from disk, but then hit some error halfway.
- Fallback to
fill_dummy_bgs()if any error gets hit inbtrfs_read_block_groups()Of course, this still needs rescue=ibadroots mount option.
With that, rescue=ibadroots can handle extent tree corruption more gracefully and allow a better recover chance.
Reported-by Zhenyu Wu <[email protected]>Signed-off-by Qu Wenruo <[email protected]>---
Changelog:
v2:
- Don't try to fill with dummy block groups when we hit ENOMEM--- fs/btrfs/block-group.c | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-)fs/btrfs/block-group.c +6
-1@@ -2105,11 +2105,16 @@ static int fill_dummy_bgs(struct btrfs_fs_info *fs_info) bg->used = em->len; bg->flags = map->type; ret =btrfs_add_block_group_cache (fs_info, bg); -if (ret) { +/* +* We may have some block groups filled already, thus ignore +* the -EEXIST error. +*/ +if (ret && ret != -EEXIST ) {btrfs_remove_free_space_cache (bg);btrfs_put_block_group (bg);break ; } + ret =0 ;btrfs_update_space_info (fs_info, bg->flags, em->len, em->len,0 ,0 , &space_info); bg->space_info = space_info;
fs/btrfs/block-group.c int btrfs_read_block_groups(struct btrfs_fs_info *info) · line 2225
Quoted 9 lines
fs/btrfs/block-group.c +6
-0@@ -2212,6 +2217,13 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info) ret =check_chunk_block_group_mappings (info); error:btrfs_free_path (path); +/* +* Either we have no extent_root (already implies IGNOREBADROOTS), or +* we hit error and have IGNOREBADROOTS, then we can try to use dummy +* bgs. +*/ +if (!info->extent_root || (ret &&btrfs_test_opt (info,IGNOREBADROOTS )))
I missed this part before, my bad.
info->extent_root can not be NULL here, which is checked at the
beginning of the function.
2138 if (!info->extent_root)
2139 return fill_dummy_bgs(info);
So should be
if (ret && btrfs_test_opt(info, IGNOREBADROOTS)) {
info->extent_root = NULL;
ret = fill_dummy_bgs(info);
}Thanks, Anand
Quoted 3 lines
fs/btrfs/block-group.c +1
-0+ ret =fill_dummy_bgs (info);return ret; }
Quoted 77 lines · Anand Jain
Quoted 64 lines · Qu Wenruo
When extent tree gets corrupted, normally it's not extent tree root, but one toasted tree leaf/node.
In that case, rescue=ibadroots mount option won't help as it can only handle the extent tree root corruption.
This patch will enhance the behavior by:
- Allow
fill_dummy_bgs()to ignore-EEXISTerrorThis means we may have some block group items read from disk, but then hit some error halfway.
- Fallback to
fill_dummy_bgs()if any error gets hit inbtrfs_read_block_groups()Of course, this still needs rescue=ibadroots mount option.
With that, rescue=ibadroots can handle extent tree corruption more gracefully and allow a better recover chance.
Reported-by Zhenyu Wu <[email protected]>Signed-off-by Qu Wenruo <[email protected]>---
Changelog:
v2:
- Don't try to fill with dummy block groups when we hit ENOMEM
---
fs/btrfs/block-group.c| 14 +++++++++++++-1 file changed, 13 insertions(+), 1 deletion(-)fs/btrfs/block-group.c +6
-1@@ -2105,11 +2105,16 @@ static int fill_dummy_bgs(struct btrfs_fs_info *fs_info) bg->used = em->len; bg->flags = map->type; ret = btrfs_add_block_group_cache(fs_info, bg); - if (ret) { + /* + * We may have some block groups filled already, thus ignore + * the -EEXIST error. + */ + if (ret && ret != -EEXIST) { btrfs_remove_free_space_cache(bg); btrfs_put_block_group(bg); break; } + ret = 0; btrfs_update_space_info(fs_info, bg->flags, em->len, em->len, 0, 0, &space_info);
bg->space_info=space_info;fs/btrfs/block-group.c +0
-0@@ -2212,6 +2217,13 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info) ret = check_chunk_block_group_mappings(info);error:
btrfs_free_path(path);
+ /*Quoted 8 lines
+ * Either we have no
extent_root(already implies IGNOREBADROOTS), or
+ * we hit error and have IGNOREBADROOTS, then we can try to use dummy
+ * bgs.
+ */
+ if (!info->extent_root || (ret &&btrfs_test_opt(info, IGNOREBADROOTS)))I missed this part before, my bad.
Quoted 5 lines
info->extent_root can not be NULL here, which is checked at the
beginning of the function.2138 if (!info->extent_root)
2139 returnfill_dummy_bgs(info);
Oh, you're right.
Quoted 4 lines
So should be
if (ret &&btrfs_test_opt(info, IGNOREBADROOTS)) {info->extent_root= NULL;
But this is not needed.
We can have extent_root if only some extent tree nodes/leaves get corrupted.
Thanks,
Qu
Quoted 12 lines
ret =
fill_dummy_bgs(info);
}Thanks, Anand
Quoted 3 lines
+ ret =
fill_dummy_bgs(info);
return ret;
}
Quoted 89 lines · Qu Wenruo
Quoted 77 lines · Anand Jain
Quoted 64 lines · Qu Wenruo
When extent tree gets corrupted, normally it's not extent tree root, but one toasted tree leaf/node.
In that case, rescue=ibadroots mount option won't help as it can only handle the extent tree root corruption.
This patch will enhance the behavior by:
- Allow
fill_dummy_bgs()to ignore-EEXISTerrorThis means we may have some block group items read from disk, but then hit some error halfway.
- Fallback to
fill_dummy_bgs()if any error gets hit inbtrfs_read_block_groups()Of course, this still needs rescue=ibadroots mount option.
With that, rescue=ibadroots can handle extent tree corruption more gracefully and allow a better recover chance.
Reported-by Zhenyu Wu <[email protected]>Signed-off-by Qu Wenruo <[email protected]>---
Changelog:
v2:
- Don't try to fill with dummy block groups when we hit ENOMEM
---
fs/btrfs/block-group.c| 14 +++++++++++++-1 file changed, 13 insertions(+), 1 deletion(-)fs/btrfs/block-group.c +6
-1@@ -2105,11 +2105,16 @@ static int fill_dummy_bgs(struct btrfs_fs_info *fs_info) bg->used = em->len; bg->flags = map->type; ret = btrfs_add_block_group_cache(fs_info, bg); - if (ret) { + /* + * We may have some block groups filled already, thus ignore + * the -EEXIST error. + */ + if (ret && ret != -EEXIST) { btrfs_remove_free_space_cache(bg); btrfs_put_block_group(bg); break; } + ret = 0; btrfs_update_space_info(fs_info, bg->flags, em->len, em->len, 0, 0, &space_info);
bg->space_info=space_info;fs/btrfs/block-group.c +0
-0@@ -2212,6 +2217,13 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info) ret = check_chunk_block_group_mappings(info);error:
btrfs_free_path(path);
+ /*Quoted 8 lines
+ * Either we have no
extent_root(already implies IGNOREBADROOTS), or
+ * we hit error and have IGNOREBADROOTS, then we can try to use dummy
+ * bgs.
+ */
+ if (!info->extent_root || (ret &&btrfs_test_opt(info, IGNOREBADROOTS)))I missed this part before, my bad.
Quoted 5 lines
info->extent_root can not be NULL here, which is checked at the beginning of the function.
2138 if (!info->extent_root)
2139 returnfill_dummy_bgs(info);Oh, you're right.
Quoted 1 line
Quoted 8 lines
Quoted 3 lines
So should be
if (ret &&btrfs_test_opt(info, IGNOREBADROOTS)) {info->extent_root= NULL;But this is not needed.
We can have
extent_rootif only some extent tree nodes/leaves get corrupted.
In fact, if here we manually set extent_root to NULL, we will leak some extent buffers.
Thanks,
Qu
Quoted 16 lines
Thanks,
QuQuoted 12 lines
ret =
fill_dummy_bgs(info);
}Thanks, Anand
Quoted 3 lines
+ ret =
fill_dummy_bgs(info);
return ret;
}
Quoted 96 lines · Qu Wenruo
Quoted 93 lines · Qu Wenruo
Quoted 77 lines · Anand Jain
Quoted 64 lines · Qu Wenruo
When extent tree gets corrupted, normally it's not extent tree root, but one toasted tree leaf/node.
In that case, rescue=ibadroots mount option won't help as it can only handle the extent tree root corruption.
This patch will enhance the behavior by:
- Allow
fill_dummy_bgs()to ignore-EEXISTerrorThis means we may have some block group items read from disk, but then hit some error halfway.
- Fallback to
fill_dummy_bgs()if any error gets hit inbtrfs_read_block_groups()Of course, this still needs rescue=ibadroots mount option.
With that, rescue=ibadroots can handle extent tree corruption more gracefully and allow a better recover chance.
Reported-by Zhenyu Wu <[email protected]>Signed-off-by Qu Wenruo <[email protected]>---
Changelog:
v2:
- Don't try to fill with dummy block groups when we hit ENOMEM
---
fs/btrfs/block-group.c| 14 +++++++++++++-1 file changed, 13 insertions(+), 1 deletion(-)fs/btrfs/block-group.c +6
-1@@ -2105,11 +2105,16 @@ static int fill_dummy_bgs(struct btrfs_fs_info *fs_info) bg->used = em->len; bg->flags = map->type; ret = btrfs_add_block_group_cache(fs_info, bg); - if (ret) { + /* + * We may have some block groups filled already, thus ignore + * the -EEXIST error. + */ + if (ret && ret != -EEXIST) { btrfs_remove_free_space_cache(bg); btrfs_put_block_group(bg); break; } + ret = 0; btrfs_update_space_info(fs_info, bg->flags, em->len, em->len, 0, 0, &space_info);
bg->space_info=space_info;fs/btrfs/block-group.c +0
-0@@ -2212,6 +2217,13 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info) ret = check_chunk_block_group_mappings(info);error:
btrfs_free_path(path);
+ /*Quoted 8 lines
+ * Either we have no
extent_root(already implies IGNOREBADROOTS), or
+ * we hit error and have IGNOREBADROOTS, then we can try to use dummy
+ * bgs.
+ */
+ if (!info->extent_root || (ret &&btrfs_test_opt(info, IGNOREBADROOTS)))I missed this part before, my bad.
Quoted 5 lines
info->extent_root can not be NULL here, which is checked at the beginning of the function.
2138 if (!info->extent_root)
2139 returnfill_dummy_bgs(info);Oh, you're right.
Quoted 4 lines
So should be
if (ret &&btrfs_test_opt(info, IGNOREBADROOTS)) {info->extent_root= NULL;
Quoted 7 lines
Quoted 4 lines
But this is not needed.
We can have
extent_rootif only some extent tree nodes/leaves get corrupted.In fact, if here we manually set
extent_rootto NULL, we will leak some extent buffers.
Right. After I replied, I thought so. Thanks for verifying.
-Anand
Quoted 19 lines
Thanks,
QuQuoted 16 lines
Thanks,
QuQuoted 12 lines
ret =
fill_dummy_bgs(info);
}Thanks, Anand
Quoted 3 lines
+ ret =
fill_dummy_bgs(info);
return ret;
}