* [PATCH 2/6] e2fsck: fix file systems with an overly large s_first_meta_bg
2014-08-09 17:10 [PATCH 1/6] libext2fs: avoid buffer overflow if s_first_meta_bg is too big Theodore Ts'o
@ 2014-08-09 17:10 ` Theodore Ts'o
2014-08-09 17:10 ` [PATCH 3/6] resize2fs: disable the meta_bg feature if necessary Theodore Ts'o
` (3 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: Theodore Ts'o @ 2014-08-09 17:10 UTC (permalink / raw)
To: Ext4 Developers List; +Cc: antymat+debian, 756922, Theodore Ts'o
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
---
e2fsck/problem.c | 5 +++++
e2fsck/problem.h | 3 +++
e2fsck/super.c | 12 ++++++++++++
3 files changed, 20 insertions(+)
diff --git a/e2fsck/problem.c b/e2fsck/problem.c
index 6d9b1af..57c2e39 100644
--- a/e2fsck/problem.c
+++ b/e2fsck/problem.c
@@ -440,6 +440,11 @@ static struct e2fsck_problem problem_table[] = {
N_("@S 64bit filesystems needs extents to access the whole disk. "),
PROMPT_FIX, PR_PREEN_OK | PR_NO_OK},
+ /* The first_meta_bg is too big */
+ { PR_0_FIRST_META_BG_TOO_BIG,
+ N_("First_meta_bg is too big. (%N, max value %g). "),
+ PROMPT_CLEAR, 0 },
+
/* Pass 1 errors */
/* Pass 1: Checking inodes, blocks, and sizes */
diff --git a/e2fsck/problem.h b/e2fsck/problem.h
index d673a4e..3426a22 100644
--- a/e2fsck/problem.h
+++ b/e2fsck/problem.h
@@ -252,6 +252,9 @@ struct problem_context {
/* 64bit is set but extents are not set. */
#define PR_0_64BIT_WITHOUT_EXTENTS 0x000048
+/* The first_meta_bg is too big */
+#define PR_0_FIRST_META_BG_TOO_BIG 0x000049
+
/*
* Pass 1 errors
*/
diff --git a/e2fsck/super.c b/e2fsck/super.c
index 8d468e6..768316a 100644
--- a/e2fsck/super.c
+++ b/e2fsck/super.c
@@ -591,6 +591,18 @@ void check_super_block(e2fsck_t ctx)
ext2fs_mark_super_dirty(fs);
}
+ if ((fs->super->s_feature_incompat & EXT2_FEATURE_INCOMPAT_META_BG) &&
+ (fs->super->s_first_meta_bg > fs->desc_blocks)) {
+ pctx.group = fs->desc_blocks;
+ pctx.num = fs->super->s_first_meta_bg;
+ if (fix_problem(ctx, PR_0_FIRST_META_BG_TOO_BIG, &pctx)) {
+ fs->super->s_feature_incompat &=
+ ~EXT2_FEATURE_INCOMPAT_META_BG;
+ fs->super->s_first_meta_bg = 0;
+ ext2fs_mark_super_dirty(fs);
+ }
+ }
+
/*
* Verify the group descriptors....
*/
--
2.0.0
^ permalink raw reply related [flat|nested] 6+ messages in thread* [PATCH 3/6] resize2fs: disable the meta_bg feature if necessary
2014-08-09 17:10 [PATCH 1/6] libext2fs: avoid buffer overflow if s_first_meta_bg is too big Theodore Ts'o
2014-08-09 17:10 ` [PATCH 2/6] e2fsck: fix file systems with an overly large s_first_meta_bg Theodore Ts'o
@ 2014-08-09 17:10 ` Theodore Ts'o
2014-08-09 17:10 ` [PATCH 4/6] tests: make sure MKE2FS_FIRST_META_BG is unset while running tests Theodore Ts'o
` (2 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: Theodore Ts'o @ 2014-08-09 17:10 UTC (permalink / raw)
To: Ext4 Developers List; +Cc: antymat+debian, 756922, Theodore Ts'o
When shrinking a file system, if the number block groups drops below
the point where we started using the meta_bg layout, disable the
meta_bg feature and set s_first_meta_bg to zero. This is necessary to
avoid creating an invalid/corrupted file system after the shrink.
Addresses-Debian-Bug: 756922
Reported-by: Marcin Wolcendorf <antymat+debian@chelmska.waw.pl>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
---
resize/resize2fs.c | 17 +++++++++++++----
1 file changed, 13 insertions(+), 4 deletions(-)
diff --git a/resize/resize2fs.c b/resize/resize2fs.c
index 9641b1e..b3755f6 100644
--- a/resize/resize2fs.c
+++ b/resize/resize2fs.c
@@ -489,6 +489,13 @@ retry:
fs->super->s_reserved_gdt_blocks = new;
}
+ if ((fs->super->s_feature_incompat & EXT2_FEATURE_INCOMPAT_META_BG) &&
+ (fs->super->s_first_meta_bg > fs->desc_blocks)) {
+ fs->super->s_feature_incompat &=
+ ~EXT2_FEATURE_INCOMPAT_META_BG;
+ fs->super->s_first_meta_bg = 0;
+ }
+
/*
* Update the location of the backup superblocks if the
* sparse_super2 feature is enabled.
@@ -998,13 +1005,15 @@ static errcode_t blocks_to_move(ext2_resize_t rfs)
ext2fs_mark_block_bitmap2(rfs->reserve_blocks, blk);
}
- if (fs->super->s_feature_incompat & EXT2_FEATURE_INCOMPAT_META_BG) {
+ if (old_fs->super->s_feature_incompat & EXT2_FEATURE_INCOMPAT_META_BG)
old_blocks = old_fs->super->s_first_meta_bg;
+ else
+ old_blocks = old_fs->desc_blocks +
+ old_fs->super->s_reserved_gdt_blocks;
+ if (fs->super->s_feature_incompat & EXT2_FEATURE_INCOMPAT_META_BG)
new_blocks = fs->super->s_first_meta_bg;
- } else {
- old_blocks = old_fs->desc_blocks + old_fs->super->s_reserved_gdt_blocks;
+ else
new_blocks = fs->desc_blocks + fs->super->s_reserved_gdt_blocks;
- }
retval = reserve_sparse_super2_last_group(rfs, meta_bmap);
if (retval)
--
2.0.0
^ permalink raw reply related [flat|nested] 6+ messages in thread* [PATCH 4/6] tests: make sure MKE2FS_FIRST_META_BG is unset while running tests
2014-08-09 17:10 [PATCH 1/6] libext2fs: avoid buffer overflow if s_first_meta_bg is too big Theodore Ts'o
2014-08-09 17:10 ` [PATCH 2/6] e2fsck: fix file systems with an overly large s_first_meta_bg Theodore Ts'o
2014-08-09 17:10 ` [PATCH 3/6] resize2fs: disable the meta_bg feature if necessary Theodore Ts'o
@ 2014-08-09 17:10 ` Theodore Ts'o
2014-08-09 17:10 ` [PATCH 5/6] tests: add f_first_meta_bg_too_big test Theodore Ts'o
2014-08-09 17:10 ` [PATCH 6/6] tests: add the r_meta_bg_shrink test Theodore Ts'o
4 siblings, 0 replies; 6+ messages in thread
From: Theodore Ts'o @ 2014-08-09 17:10 UTC (permalink / raw)
To: Ext4 Developers List; +Cc: antymat+debian, 756922, Theodore Ts'o
If the developer has set the MKE2FS_FIRST_META_BG environment
variable, this can cause test failures.
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
---
tests/test_config | 1 +
1 file changed, 1 insertion(+)
diff --git a/tests/test_config b/tests/test_config
index 6789656..1872e5c 100644
--- a/tests/test_config
+++ b/tests/test_config
@@ -32,6 +32,7 @@ E2FSCK_CONFIG=/dev/null
export E2FSCK_CONFIG
MKE2FS_CONFIG=./mke2fs.conf
export MKE2FS_CONFIG
+unset MKE2FS_FIRST_META_BG
E2FSPROGS_SKIP_PROGRESS=yes
export E2FSPROGS_SKIP_PROGRESS
EXT2FS_NO_MTAB_OK=yes
--
2.0.0
^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH 5/6] tests: add f_first_meta_bg_too_big test
2014-08-09 17:10 [PATCH 1/6] libext2fs: avoid buffer overflow if s_first_meta_bg is too big Theodore Ts'o
` (2 preceding siblings ...)
2014-08-09 17:10 ` [PATCH 4/6] tests: make sure MKE2FS_FIRST_META_BG is unset while running tests Theodore Ts'o
@ 2014-08-09 17:10 ` Theodore Ts'o
2014-08-09 17:10 ` [PATCH 6/6] tests: add the r_meta_bg_shrink test Theodore Ts'o
4 siblings, 0 replies; 6+ messages in thread
From: Theodore Ts'o @ 2014-08-09 17:10 UTC (permalink / raw)
To: Ext4 Developers List; +Cc: antymat+debian, 756922, Theodore Ts'o
The test verifies that e2fsck can properly fix a file system where the
value of s_first_meta_bg in the superblock is larger than the number
of block group descriptors in the file system. E2fsck will fix this
by clearing the meta_bg feature.
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
---
tests/f_first_meta_bg_too_big/expect.1 | 20 ++++++++++++++++++++
tests/f_first_meta_bg_too_big/expect.2 | 7 +++++++
tests/f_first_meta_bg_too_big/image.gz | Bin 0 -> 533 bytes
tests/f_first_meta_bg_too_big/name | 1 +
4 files changed, 28 insertions(+)
create mode 100644 tests/f_first_meta_bg_too_big/expect.1
create mode 100644 tests/f_first_meta_bg_too_big/expect.2
create mode 100644 tests/f_first_meta_bg_too_big/image.gz
create mode 100644 tests/f_first_meta_bg_too_big/name
diff --git a/tests/f_first_meta_bg_too_big/expect.1 b/tests/f_first_meta_bg_too_big/expect.1
new file mode 100644
index 0000000..85786bd
--- /dev/null
+++ b/tests/f_first_meta_bg_too_big/expect.1
@@ -0,0 +1,20 @@
+First_meta_bg is too big. (2, max value 1). Clear? yes
+
+Pass 1: Checking inodes, blocks, and sizes
+Pass 2: Checking directory structure
+Pass 3: Checking directory connectivity
+Pass 4: Checking reference counts
+Pass 5: Checking group summary information
+Block bitmap differences: -3
+Fix? yes
+
+Free blocks count wrong for group #0 (79, counted=80).
+Fix? yes
+
+Free blocks count wrong (79, counted=80).
+Fix? yes
+
+
+test_filesys: ***** FILE SYSTEM WAS MODIFIED *****
+test_filesys: 11/16 files (0.0% non-contiguous), 20/100 blocks
+Exit status is 1
diff --git a/tests/f_first_meta_bg_too_big/expect.2 b/tests/f_first_meta_bg_too_big/expect.2
new file mode 100644
index 0000000..44bff62
--- /dev/null
+++ b/tests/f_first_meta_bg_too_big/expect.2
@@ -0,0 +1,7 @@
+Pass 1: Checking inodes, blocks, and sizes
+Pass 2: Checking directory structure
+Pass 3: Checking directory connectivity
+Pass 4: Checking reference counts
+Pass 5: Checking group summary information
+test_filesys: 11/16 files (0.0% non-contiguous), 20/100 blocks
+Exit status is 0
diff --git a/tests/f_first_meta_bg_too_big/image.gz b/tests/f_first_meta_bg_too_big/image.gz
new file mode 100644
index 0000000000000000000000000000000000000000..527937040de93a57428e73c8aee9181212b5efc4
GIT binary patch
literal 533
zc-oWi=HNKx`z)A=IWspgJ(c0@-Tm3Zfiesq%BLTb_idY2_ClrAs>5kT^Xe-*MGpFu
z3A8^_NNQiD!NJBV{_;V}+`q;xmyW9L_#wVSR6H=?qU*WozZt7~tu&-2ooC<ov*_gR
z??uyp@AGEb*}~wnYNmv=;Vi>*(=C7QJD4PB{4}XPcfYjLw9@;kmAjL62CTi_b+Nlp
z=~$~ti?yNst95qu-%_4_KHPnL+xfWI|G(_Fea+jp`|rz-r~X{N`Bj`hE;dTa=6+ev
zmTj{2!A4K^t&~WYsp@GsxY0uXaYfw!vwOs}&ThE6y*=U2T)pD_@1^s~^XAr6{=avy
z?DW<-f0o@3Tl;_2{i8*5Za*!WT7T{T?@fQ7#!Y;l)gS)%{nWRwrk#p8o4+|Y@x_l)
z$)2r?6BpWC+kZpSAn?*3<AO<Lr;3xUmt-yvp0oYF$+FAYv!l)ZVsGxLtUX+<B4NHZ
z>Ti9H$X9-!3)34Kez9M%U+|CdEB}gmhri6P<OBYLg%}tNiZ>W9FTFiid*%1uR~^
zf7|`FkK+4&e~US>-oC8hkK4ccBZ@L7^OxJJ*vBl>>Ad)56JvcszQKPT{@L}HHhk#*
axm@<~guCnvlz^Uv`G<RLRx?asWB>r>%kgsn
literal 0
Hc-jL100001
diff --git a/tests/f_first_meta_bg_too_big/name b/tests/f_first_meta_bg_too_big/name
new file mode 100644
index 0000000..47ec96d
--- /dev/null
+++ b/tests/f_first_meta_bg_too_big/name
@@ -0,0 +1 @@
+s_first_meta_bg is too large
--
2.0.0
^ permalink raw reply related [flat|nested] 6+ messages in thread* [PATCH 6/6] tests: add the r_meta_bg_shrink test
2014-08-09 17:10 [PATCH 1/6] libext2fs: avoid buffer overflow if s_first_meta_bg is too big Theodore Ts'o
` (3 preceding siblings ...)
2014-08-09 17:10 ` [PATCH 5/6] tests: add f_first_meta_bg_too_big test Theodore Ts'o
@ 2014-08-09 17:10 ` Theodore Ts'o
4 siblings, 0 replies; 6+ messages in thread
From: Theodore Ts'o @ 2014-08-09 17:10 UTC (permalink / raw)
To: Ext4 Developers List; +Cc: antymat+debian, 756922, Theodore Ts'o
This test checks to make sure resize2fs can properly handle a file
system which started life as a normal ext4 file system and then was
grown to a size where meta_bg was enabled, and then shrunk back below
the point where the meta_bg format is still needed.
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
---
tests/r_meta_bg_shrink/script | 34 ++++++++++++++++++++++++++++++++++
1 file changed, 34 insertions(+)
create mode 100644 tests/r_meta_bg_shrink/script
diff --git a/tests/r_meta_bg_shrink/script b/tests/r_meta_bg_shrink/script
new file mode 100644
index 0000000..a29c28a
--- /dev/null
+++ b/tests/r_meta_bg_shrink/script
@@ -0,0 +1,34 @@
+if test -x $RESIZE2FS_EXE -a -x $DEBUGFS_EXE; then
+
+test_description="meta_bg shrink"
+FEATURES="-t ext4 -O 64bit,meta_bg,^resize_inode -b 1024"
+SIZE_1=1G
+SIZE_2=48M
+LOG=$test_name.log
+E2FSCK=../e2fsck/e2fsck
+RESIZE2FS_OPTS=-f
+
+. $cmd_dir/scripts/resize_test
+
+export MKE2FS_FIRST_META_BG=2
+resize_test
+unset MKE2FS_FIRST_META_BG
+
+RC=$?
+if [ $RC -eq 0 ]; then
+ echo "$test_name: $test_description: ok"
+ touch $test_name.ok
+elif [ $RC -eq 111 ]; then
+ echo "$test_name: $test_description: skipped"
+ touch $test_name.ok
+else
+ echo "$test_name: $test_description: failed"
+ ln $LOG $test_name.failed
+fi
+
+unset FEATURES SIZE_1 SIZE_2 LOG E2FSCK
+
+else #if test -x $RESIZE2FS_EXE -a -x $DEBUGFS_EXE; then
+ echo "$test_name: $test_description: skipped"
+fi
+
--
2.0.0
^ permalink raw reply related [flat|nested] 6+ messages in thread