The quota_handle wasn't getting closed in quota_compare_and_update().
Fix this, and also make sure that quota_file_close() doesn't
unnecessarily modify the quota inode if it's not necessary. Otherwise
e2fsck will claim that the file system is modified when it didn't need
to be.
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Reviewed-by: Aditya Kali <adityakali@google.com>
Previously if there was a missing quota entry --- i.e., if there were
files owned by group "eng", but there was no quota record for group
"eng", e2fsck would not notice the missing entry. This means that the
usage informtion would not be properly repaired. This is unfortunate.
Fix this by marking each quota record in quota_dict that has a
corresponding record on disk, and then check to see if there are any
records in quota_dict that have not been marked as having been seen.
In that case, we know we need to update the relevant quota inode.
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Reviewed-by: Aditya Kali <adityakali@google.com>
In scan_dquota_callback() we were copying dqb_off from the on-disk
dquot structure to the dqot structure that we have constructed in
memory. This is a bad idea, because if we detect that the quota inode
is corrupted and needs to be replaced, when we later write out the
inode, the fact that we have a non-zero dqb_off value means that quota
routines will not bother updating the quota tree index, since
presumably the quota tree index already has an entry for this dqot.
The problem is that e2fsck, the only user of these functions, has
zapped the quota inode so it can be written from scratch. So this
means the index for the quota records are not written out, so the
kernel can not find any of the records in the quota inodes. Oops.
The fix is simple; there is no reason to copy the dqb_off field into
the quota_dict copy of struct dquot.
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Reviewed-by: Aditya Kali <adityakali@google.com>
If the user passes the -O option to logdump, try to dump old log
contents. This can be used to try to track down journal problems even
after the journal has been replayed.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
The logdump command doesn't know how to deal with revoke tables in
64bit journals, so teach it to do this.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
The combination of 128-byte inodes and inline_data is silly, since
there's no room in the inode table. Unfortunately, if neither
mke2fs.conf nor the mkfs command line options specify an inode size,
the default inode size is set to 128 bytes (by libext2fs) and the
warning isn't printed. Therefore, always do the check-and-warning.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Since the code to copy-in a socket when creating a filesystem is
fairly simple, just do it here.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Fix various small resource leaks and error code handling issues that
Coverity pointed out.
Fixes-Coverity-Bugs: 11919{39-45}, 1174118, 1049160, 1049144
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Fix some minor bugs relating to passing CFLAGS to cppcheck, and
package the cppcheck output into nicer looking reports.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
dict_uint_cmp() returns an usigned int value in int type, which
could mess the dict key comparison when the difference of two
keys is greater than INT_MAX.
Signed-off-by: Niu Yawei <yawei.niu@intel.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Adding the pkgconfigdir variable allows specifying an installation
location for pkg-config files independent of libdir.
Signed-off-by: David Michael <fedora.dm0@gmail.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
The checking of types in parse-types.sh doesn't make much sense in a
cross-compilation environment, because the generated binary is
executed on build machine.
So even if asm_types.h has got correct statements for types, it's
possible that the generated binary will report an error, because these
types are for the target machine.
Signed-off-by: Chen Qi <Qi.Chen@windriver.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
This supercedes the "whole disk" check, since it does a better job and
there are times when it is quite legitimate to want to use the whole
disk.
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
The basic idea is to provide a bit more context in this situation:
% ./misc/mke2fs -t ext4 /dev/sdc3
mke2fs 1.42.9 (4-Feb-2014)
/dev/sdc3 contains a ext4 file system
Proceed anyway? (y,n)
... by adding this bit of context:
% ./misc/mke2fs -t ext4 /dev/sdc3
mke2fs 1.42.9 (4-Feb-2014)
/dev/sdc3 contains a ext4 file system
last mounted on /SOX-backups on Mon May 5 08:59:53 2014
Proceed anyway? (y,n)
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
We've added the ability to automatically recreate a file if it doesn't
exist prior to creating the file system, since this is often used (for
example) when managing file system images for use in virtual machines.
We should at least notify the user that this is going on to avoid
surprises in the case of misspelled device/file names.
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
In the case where the new location of the inode table is before the
old inode table, the optimization which tries to optimize zero block
moves breaks. Fix it.
This fixes a bug that was tickled by the reproduction described in the
previous commit.
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
If the previous block group's inode table ends at the very end of file
system, wrap around to the beginning of the flex_bg.
This fixes a bug was tickled by:
mke2fs.conf:
frontload = {
features = extent,huge_file,flex_bg,uninit_bg,dir_nlink,extra_isize,^resize_inode,sparse_super2
hash_alg = half_md4
num_backup_sb = 0
packed_meta_blocks = 1
inode_ratio = 4194304
flex_bg_size = 262144
}
mke2fs -T frontload /tmp/foo.img 2T
resize2fs -M /tmp/foo.img
resize2fs -M /tmp/foo.img
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
The previous commit exposed bugs in the calculation for flex_bg file
systems. The problem is that since (by default) we keep the metadata
blocks for the flex_bg in the first block group of the flex_bg, and
because we don't want to overwrite metadata blocks used by the
original file system with data blocks make life easier in case the
resize is aborted for some reason, we need to treat all of the
metadata blocks in the existing flex_bg has in use for the purposes of
calculate_minimum_resize_size().
Even though this means we need to reserve more data blocks to avoid
running out of space, the net result of these two commits is a net
savings in how much we can shrink a file system.
Using the following test sequence:
mke2fs -F -t ext4 /tmp/foo.img 2T
resize2fs -M /tmp/foo.img
resize2fs -M /tmp/foo.img
resize2fs -M /tmp/foo.img
Here is the comparison in the resulting file systems between the old
and new resize2fs (units are in 4k blocks):
resize #1 resize #2 resize #3
old resize2fs 1117186 45679 43536
new resize2fs 48784 37413 37392
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
For extent-mapped file systems, we need to reserve some extra space in
case we need to grow the extent tree. Calculate the safety margin
more intelligently, so we don't overestimate the amount of space
required.
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Reported-by: Dmitry Monakhov <dmonakhov@openvz.org>
If there are any PREEN_OK problems fixed in check_super_block(), don't
skip checking the full file system.
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Filefrag doesn't catch and print the shared extent flag. Add this for
users of filefrag on file systems with shared extents (such as btrfs).
Signed-off-by: Mark Fasheh <mfasheh@suse.de>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Warn the system administrator if there is an existing file system on
the block device, and give the administrator an opportunity to abort
the mkfs operation.
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
If mke2fs needs to ask the user for permission, and the user doesn't
type anything the specified delay in the /etc/mke2fs.conf file,
proceed as if the user had said yes. The default is to do what we
currently do, which is to wait until the user answers the question one
way or the other.
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Don't ask the user if it's OK that a regular file is smaller than the
requested size. This test only makes sense if we are creating the
file system on a block device. This allow users to not need to
manually answer the "proceed?" question when creating a file system
backed by a simple file.
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Move the call to proceed_question() from check_plausibility() to its
caller. This allows more fine grained control by mke2fs about when it
might want to call check_plausibility().
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Very often people are creating file systems using regular files, so we
shouldn't ask the user to confirm using the proceed question.
Otherwise it encourages users to use the -F flag, which is a bad
thing.
We do need to continue to check if the external journal device is a
block device.
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
We already skip the low dtime check if the number of inods is greater
than the last mount or last written time. However, if a very large
file system is resized sufficiently large that the number of inodes is
greater than when the file system was original created, we can end up
running afoul of the low dtime check. This results in a large number
of false positives which e2fsck can fix up without causing any
problems, but it can induce a large amount of anxiety for the system
administrator.
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Reported-by: Patrik Horník <patrik@hornik.sk>
Old distros may not have the "truncate" tool, so use "dd" instead.
If tmpfs cannot handle a 2GB temp file (e.g. old RHEL5 and SLES 11
kernels) then skip the test instead of failing it. If this fails,
try to report better error messages instead of failing silently.
Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Fix various unused variable and use-uninitialized warnings.
Add generated files into .gitignore.
Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
We need return retval when "mke2fs -d" failed, otherwise the "$?"
would be 0 which is misleading.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Reviewed-by: Zheng Liu <wenqing.lz@taobao.com>
This allows e4defrag to work with 64-bit and bigalloc file systems.
Signed-off-by: Jon Ernst <jonernst07@gmail.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Fix the following build errors on bigendian hosts.
- ctx is a pointer, use '->' not '.'
- add missing argument to ext2fs_dirent_swab_in2
make[2]: Entering directory `/root/e2fsprogs/lib/ext2fs'
CC inline_data.c
inline_data.c: In function ‘ext2fs_inline_data_dir_iterate’:
inline_data.c:221:5: error: request for member ‘errcode’ in something not a structure or union
ctx.errcode = ext2fs_dirent_swab_in2(fs, ctx->buf, ctx->buflen, 0);
^
inline_data.c:222:9: error: request for member ‘errcode’ in something not a structure or union
if (ctx.errcode) {
^
inline_data.c: In function ‘ext2fs_inline_data_dir_expand’:
inline_data.c:364:2: error: too few arguments to function ‘ext2fs_dirent_swab_in2’
retval = ext2fs_dirent_swab_in2(fs, buf, size);
^
In file included from inline_data.c:19:0:
ext2fs.h:1569:18: note: declared here
extern errcode_t ext2fs_dirent_swab_in2(ext2_filsys fs, char *buf, size_t size,
^
make[2]: *** [inline_data.o] Error 1
Signed-off-by: Eryu Guan <guaneryu@gmail.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Reviewed-by: Zheng Liu <wenqing.lz@taobao.com>
When pass1 finds blocks that are mapped to multiple files, it will
print every duplicated block. If there are long sequences of
duplicate blocks (e.g. the e_pblk field is wrong in an extent), this
can cause a gigantic flood of output when a range could convey the
same information. Therefore, teach pass1b to print ranges when
possible.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
In ext2fs_extent_set_bmap() and ext2fs_punch_extent(), fix the parents
when altering either end of an extent so that the parent nodes reflect
the added mapping.
There's a slight complication to using fix_parents: if there are two
mappings to an lblk in the tree, the value of handle->path->curr can
point to either extent afterwards), which is documented in a comment.
Some additional color commentary from Darrick:
In the _set_bmap() case, I noticed that the "remapping last block in
extent" case would produce symptoms if we are trying to remap a
block from "extent" to "next_extent", and the two extents are
pointed to by different index nodes. _extent_replace(...,
next_extent) updates e_lblk in the leaf extent, but because there's
no _extent_fix_parents() call, the index nodes never get updated.
In the _punch_extent() case, we conclude that we need to split an
extent into two pieces since we're punching out the middle. If the
extent is the last extent in the block, the second extent will be
inserted into a new leaf node block. Without _fix_parents(), the
index node doesn't seem to get updated.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
In ext2fs_extent_free(), h(andle)->max_depth is used as a loop
conditional variable to free all the h->path[].buf pointers. However,
ext2fs_extent_delete() sets max_depth = 0 if we've removed everything
from the extent tree, which causes a subsequent _free() to leak some
buf pointers. max_depth can be re-incremented when splitting extent
nodes, but there's no guarantee that it'll reach the old value before
the free.
Therefore, remember the size of h->paths[] separately, and use that
when freeing the extent handle.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Fix a few minor bugs that cppcheck complained about.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
In ext2fs_block_alloc_stats_range(), the quantity "-inuse * n" is
calculated as a signed 32-bit quantity. Unfortunately, gcc (4.6.3 on
Ubuntu 12.04) doesn't sign-extend this quantity to fill the blk64_t
parameter that ext2fs_free_blocks_count_add() wants, so the end result
is that the superblock gets a ridiculously huge free block count.
Changing the declaration of 'n' to blk64_t seems to fix this.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Fix a number of things that cppcheck complains about. Most of these
are minor resource leaks and forgotten declarations.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
When we're looking for directory blocks for the inode remapping step,
we need to include inline_data directories in the remap process.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
In ext2fs_inline_data_dir_iterate(), we must be very careful to undo
any modifications we make to the dir_context pointer passed in by the
caller, because it's entirely possible that the caller will still want
to do something with the ctx or something inside.
Specifically, ext2fs_dblist_dir_iterate() wants to be able to free
ctx->buf, and it reuses the ctx for multiple dblist entries. That
means that assigning ctx->buf will cause weird crashes at the end of
dir_iterate().
Since we're being careful with ctx, we might as well handle adding the
INLINE_DATA flag to ctx->flags for ext2fs_process_dir_block, since the
dblist caller forgets to unset the flag before reusing the ctx.
This fixes some crashes and valgrind complaints in resize2fs, and is
necessary for the next patch, which fixes resize2fs not to corrupt
inline_data filesystems.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
When expanding an inline data inode, it's possible that the reduction
in the size of the EA structures causes the freeing of the EA block,
which changes the inode. If this happens, the local version of the
inode that ext2fs_inline_data_expand was modifying will be out of sync
with what's on the disk. This local copy gets written out to disk
after a block allocation, at which point it's possible that the inode
EA block and logical block zero point to the same physical block,
which is bad news.
Therefore, write the local copy to disk before removing the inline
data EA, and reread it afterwards.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>