前言
异常掉电情况下，如果硬件保护不周，可能导致文件系统损坏。如何快速的判断某个EXT4 fs是否存在错误？
当然e2fsck -p 是一个方法，但是这种方法有两个问题：
- 不得不umount文件系统
- 如果存在文件系统错误，e2fsck 会修复之
如果仅仅是简单的判断某个线上文件系统是否存在错误，有无轻量级的方法呢？ tune2fs 来帮忙
方法
tune2fs -l /dev/sdX
这个命令可以帮忙检测EXT4文件系统是否存在文件系统错误。
输出如下：
root@scalars04:/var/log# tune2fs -l /dev/mapper/3600d02310004a2152807e4a4208b3006-part2
tune2fs 1.42 (29-Nov-2011)
Filesystem volume name: <none>
Last mounted on: /data/osd.66
Filesystem UUID: 0a6eb642-a9f3-4f76-9f2d-b6aaf12605b9
Filesystem magic number: 0xEF53
Filesystem revision #: 1 (dynamic)
Filesystem features: has_journal ext_attr dir_index filetype needs_recovery extent 64bit flex_bg sparse_super huge_file uninit_bg dir_nlink extra_isize
Filesystem flags: signed_directory_hash
Default mount options: user_xattr acl
Filesystem state: clean with errors
Errors behavior: Continue
Filesystem OS type: Linux
Inode count: 615763968
Block count: 9852209915
Reserved block count: 0
Free blocks: 3533386093
Free inodes: 614654549
First block: 0
Block size: 4096
Fragment size: 4096
Blocks per group: 32768
Fragments per group: 32768
Inodes per group: 2048
Inode blocks per group: 128
Flex block group size: 16
Filesystem created: Mon Jan 18 20:47:21 2016
Last mount time: Tue Nov 29 14:53:52 2016
Last write time: Fri Dec 16 03:35:36 2016
Mount count: 13
Maximum mount count: -1
Last checked: Mon Jan 18 20:47:21 2016
Check interval: 0 (<none>)
Lifetime writes: 27 TB
Reserved blocks uid: 0 (user root)
Reserved blocks gid: 0 (group root)
First inode: 11
Inode size: 256
Required extra isize: 28
Desired extra isize: 28
Journal inode: 8
Default directory hash: half_md4
Directory Hash Seed: 4166ec4b-4463-4c13-9395-23c1a7a0bc5b
Journal backup: inode blocks
FS Error count: 233
First error time: Tue May 24 10:29:21 2016
First error function: ext4_mb_generate_buddy
First error line #: 756
First error inode #: 0
First error block #: 0
Last error time: Fri Dec 16 03:35:36 2016
Last error function: ext4_mb_generate_buddy
Last error line #: 757
Last error inode #: 0
Last error block #: 0
root@scalars04:/var/log#
原理
tune2fs属于 e2fsprogs 这个deb package：
root@BEAN-3:~# which tune2fs
/sbin/tune2fs
root@BEAN-3:~# dpkg -S /sbin/tune2fs
e2fsprogs: /sbin/tune2fs
从代码上看：
void list_super (struct ext2_super_block * s)
{
list_super2(s, stdout);
}
void list_super2(struct ext2_super_block * sb, FILE *f)
{
...
if (sb->s_first_error_time) {
tm = sb->s_first_error_time;
fprintf(f, "First error time: %s", ctime(&tm));
memset(buf, 0, sizeof(buf));
strncpy(buf, (char *)sb->s_first_error_func,
sizeof(sb->s_first_error_func));
fprintf(f, "First error function: %s\n", buf);
fprintf(f, "First error line #: %u\n",
sb->s_first_error_line);
fprintf(f, "First error inode #: %u\n",
sb->s_first_error_ino);
fprintf(f, "First error block #: %llu\n",
sb->s_first_error_block);
}
if (sb->s_last_error_time) {
tm = sb->s_last_error_time;
fprintf(f, "Last error time: %s", ctime(&tm));
memset(buf, 0, sizeof(buf));
strncpy(buf, (char *)sb->s_last_error_func,
sizeof(sb->s_last_error_func));
fprintf(f, "Last error function: %s\n", buf);
fprintf(f, "Last error line #: %u\n",
sb->s_last_error_line);
fprintf(f, "Last error inode #: %u\n",
sb->s_last_error_ino);
fprintf(f, "Last error block #: %llu\n",
sb->s_last_error_block);
}
...
}
内核代码中fs/ext4/super.c中有__save_error_info 函数会将错误信息记录在superblock：
static void __save_error_info(struct super_block *sb, const char *func,
unsigned int line)
{
struct ext4_super_block *es = EXT4_SB(sb)->s_es;
EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS;
if (bdev_read_only(sb->s_bdev))
return;
es->s_state |= cpu_to_le16(EXT4_ERROR_FS);
es->s_last_error_time = cpu_to_le32(get_seconds());
strncpy(es->s_last_error_func, func, sizeof(es->s_last_error_func));
es->s_last_error_line = cpu_to_le32(line);
if (!es->s_first_error_time) {
es->s_first_error_time = es->s_last_error_time;
strncpy(es->s_first_error_func, func,
sizeof(es->s_first_error_func));
es->s_first_error_line = cpu_to_le32(line);
es->s_first_error_ino = es->s_last_error_ino;
es->s_first_error_block = es->s_last_error_block;
}
/*
* Start the daily error reporting function if it hasn't been
* started already
*/
if (!es->s_error_count)
mod_timer(&EXT4_SB(sb)->s_err_report, jiffies + 24*60*60*HZ);
le32_add_cpu(&es->s_error_count, 1);
}
会调用该函数的场景比较多，就不一一赘述了。