1

tl; dr:我想查看/获取 dmesg 上报告的块号所属的文件名,vm.block_dump=1例如 dmesg:bash(13515): READ block 5434824 on xvda3 (32 sectors)

当例如。sudo sysctl -w vm.block_dump=1或例如。echo '1' | sudo tee /proc/sys/vm/block_dump然后“Linux 报告发生的所有磁盘读取和写入操作,以及对文件进行的所有块污染。[...] block_dump 的输出被写入内核输出,并且可以使用“dmesg”检索它。当你使用 block_dump 并且您的内核日志记录级别还包括内核调试消息,您可能希望关闭 klogd,否则将记录 block_dump 的输出,导致磁盘活动不正常。” (从这里引用)

“块脏污”不是问题,例如。

[ 3140.559675] systemd-journal(291): dirtied inode 399135 (system.journal) on xvda3

我可以看到它的文件名是这样的:

$ echo -e 'open /dev/xvda3\n ncheck 399135' | sudo debugfs -f -
debugfs 1.44.2 (14-May-2018)
debugfs: open /dev/xvda3
debugfs:  ncheck 399135
Inode   Pathname
399135  /var/log/journal/12c5e521101c444594b96b53751551a8/system.journal

问题在于“Linux 报告发生的所有磁盘读写操作”(引用上面的内容),因为它们是以块的形式报告的,例如。

[ 3140.376827] kworker/u24:3(21616): WRITE block 11037768 on xvda3 (8 sectors)
[ 3140.724725] bash(13515): READ block 5434824 on xvda3 (32 sectors)
[ 3140.725483] date(13515): READ block 5434896 on xvda3 (160 sectors)
[ 3140.728946] sed(13519): READ block 5143680 on xvda3 (32 sectors)
[ 3140.736022] sleep(13522): READ block 5379184 on xvda3 (24 sectors)
[ 3140.804803] qubes-gui(522): READ block 5179952 on xvda3 (16 sectors)
[ 3140.806519] Xorg(599): READ block 7420192 on xvda3 (176 sectors)
[ 3140.810348] InputThread(613): READ block 7418560 on xvda3 (112 sectors)
[ 3140.815866] at-spi2-registr(812): READ block 5654512 on xvda3 (8 sectors)
[ 3140.816860] xdg-desktop-por(888): READ block 5795168 on xvda3 (8 sectors)
[ 3140.818716] gnome-terminal-(865): READ block 5804672 on xvda3 (16 sectors)
[ 3141.064524] sed(13531): READ block 3446048 on xvda3 (16 sectors)
[ 3141.130808] systemd(571): READ block 4744136 on xvda3 (184 sectors)

可以在此处查看负责显示此类消息的内核代码: https ://elixir.bootlin.com/linux/v4.18.5/source/block/blk-core.c#L2542

这些块都不会产生任何 inode 编号,使用以下方法:

$ echo -e 'open /dev/xvda3\n icheck 11037768' |sudo debugfs -f -
debugfs 1.44.2 (14-May-2018)
debugfs: open /dev/xvda3
debugfs:  icheck 11037768
Block   Inode number
11037768    <block not found>

我应该得到一个 inode 编号,而不是<block not found>上面的,然后我可以将其与之前的 echo 一起使用来查看文件名。

我尝试了什么:以防块号是扇区大小(512字节)的倍数并且我知道ext4块大小是4096字节(sudo blockdev --getbsz /dev/xvda),我也尝试过(使用乘法的结果):11037768*2, 11037768*4和 11037768*8 结果相同:<block not found>

我在这里想念什么?这些块是指向目录条目还是不是文件名的东西?即便如此,debugfs 不应该仍然产生一个 inode 号吗?
有没有更好的方法来获取文件名?

编辑:(icheck内部debugfs)期望的块号是类似于 debugfsstat在下报告的数字EXTENTS,例如,范围内的任何数字2172716-2172721如下所示:

$ sudo debugfs -R "stat /usr/lib/python2.7/site-packages/salt/modules/zonecfg.py" /dev/xvda3
debugfs 1.44.2 (14-May-2018)
Inode: 550529   Type: regular    Mode:  0644   Flags: 0x80000
Generation: 1781055959    Version: 0x00000000:00000001
User:     0   Group:     0   Project:     0   Size: 22179
File ACL: 0
Links: 1   Blockcount: 48
Fragment:  Address: 0    Number: 0    Size: 0
 ctime: 0x5b6ec29d:1b2e0200 -- Sat Aug 11 13:03:57 2018
 atime: 0x5b33b5a9:00000000 -- Wed Jun 27 18:04:57 2018
 mtime: 0x5b33b5a9:00000000 -- Wed Jun 27 18:04:57 2018
crtime: 0x5b6ec29d:1af0f900 -- Sat Aug 11 13:03:57 2018
Size of extra inode fields: 32
EXTENTS:
(0-5):2172716-2172721

其他方式:

$ filefrag -s -v /usr/lib/python2.7/site-packages/salt/modules/zonecfg.py
Filesystem type is: ef53
File size of /usr/lib/python2.7/site-packages/salt/modules/zonecfg.py is 22179 (6 blocks of 4096 bytes)
 ext:     logical_offset:        physical_offset: length:   expected: flags:
   0:        0..       5:    2172716..   2172721:      6:             last,eof
/usr/lib/python2.7/site-packages/salt/modules/zonecfg.py: 1 extent found

现在问题仍然存在:这些设备块号(在 dmesg 中为 xvda3 报告的块号)与这些物理偏移量有何关系?

EDIT2:我刚刚确认这些物理偏移编号与块设备编号相同(只是,显然与 dmesg 上报告的不同);下面显示了上述文件的最后一个块,我可以确认它与查看文件时相同vim

$ sudo dd bs=4096 skip=2172721 count=1 if=/dev/xvda3 | hexdump -C

我已经在 Qubes OS R4.0 Fedora 28 AppVM 内的 4.18.5 内核下对此进行了测试。.config(如有必要,我可以使用自定义 /patches 重新编译自定义内核- 欢迎提出建议)

4

1 回答 1

4

dmesg 上报告的块号vm.block_dump=1是扇区号 -512每个块的字节数 - 但是debugfs命令期望其icheck命令的块号通常4096是 ext4 的每个块的字节数,因此所需的只是除以8.

虽然块大小通常为 4096 字节,但您必须确定,所以像这样获取块大小:

$ sudo blockdev --getbsz /dev/xvda3
4096

虽然此值通常为 4096,但在某些情况下可能会有所不同,例如。mkfs.ext4with-b可以创建具有不同块大小的 ext4 文件系统;vfat 有 512 字节的块大小。

要从块大小中一次性获取除法器的分隔符(在 bash 中):

$ echo $(( `sudo blockdev --getbsz /dev/xvda3` / 512 ))
8

让我们使用这一行作为输入:
[ 3140.736022] sleep(13522): READ block 5379184 on xvda3 (24 sectors)

icheck通过将上述除以得到块号8

$ bc -l
bc 1.06.95
Copyright 1991-1994, 1997, 1998, 2000, 2004, 2006 Free Software Foundation, Inc.
This is free software with ABSOLUTELY NO WARRANTY.
For details type `warranty'. 
5379184/8
672398.00000000000000000000

我曾经bc -l确保我没有打错数字(如果它以.00000000000000000000我很可能没有)

通过使用获取路径名debugfs

$ sudo debugfs
debugfs 1.44.2 (14-May-2018)
debugfs:  open /dev/xvda3
debugfs:  icheck 672398
Block   Inode number
672398  134571
debugfs:  ncheck 134571
Inode   Pathname
134571  /usr/bin/sleep
debugfs:  close
debugfs:  quit

我不确定 dmesg 中报告的块号是否可以是每个扇区 512 字节以外的值的倍数。例如,如果底层磁盘物理上每个扇区有 4096 个字节,它们是否仍然是每个块 512 个字节?如果我猜的话,我会说假设它们总是每块 512 字节是安全的。

这是上面的脚本版本:

#!/bin/bash

#./showblock rev.02 rewritten for question/answer from: https://stackoverflow.com/q/52058914/10239615

#----
bytes_per_sector=512 #assumed that dmesg block numbers are 512 bytes each (ie. 512 bytes per sector; aka block size is 512)!
#----

#use `sudo` only when not already root
if test "`id -u`" != "0"; then
    sudo='sudo'
else
    sudo=''
fi

if ! test "$#" -ge "2"; then
  echo "Usage: $0 <device> <dmesgblocknumber> [dmesgblocknumber ...]"
  echo "Examples:"
  echo "$0 /dev/xvda3 5379184"
  echo "$0 /dev/xvda3 5379184 5129952 7420192"
  exit 1
fi

within_exit() {
  echo -e "\nSkipped current instruction within on_exit()"
}
on_exit() {
  #trap - EXIT SIGINT SIGQUIT SIGHUP  #will exit by skipping the rest of all instrunction from on_exit() eg. if C-c
  trap within_exit EXIT SIGINT SIGQUIT SIGHUP #skip only current instruction from on_exit() eg. when C-c is pressed
  #echo "first sleep"
  #sleep 10
  #echo "second sleep"
  #sleep 10
  if test "${#remaining_args[@]}" -gt 0; then
    echo -n "WARNING: There are '${#remaining_args[@]}' remaining args not processed, they are: " >&2
    for i in `seq 0 1 "$(( "${#remaining_args[@]}" - 1 ))"`; do  #seq is part of coreutils package
      echo -n "'${remaining_args[${i}]}' " >&2
    done
    echo >&2
  fi
}

trap on_exit EXIT SIGINT SIGQUIT SIGHUP

dev="$1"
shift 1

if test -z "$dev" -o ! -b "$dev"; then
  echo "Bad device name or not a device: '$dev'" >&2
  exit 1
fi

blocksize="`$sudo blockdev --getbsz "$dev"`"
if test "${blocksize:-0}" -le "0"; then #handles empty arg too
  echo "Failed getting block size for '$dev', got '$blocksize'" >&2
  exit 1
fi
#TODO: check and fail if not a multiplier
divider="$(( $blocksize / $bytes_per_sector ))"
if ! test "${divider:-0}" -gt "0"; then
  echo "Failed computing divider from: '$blocksize' / '$bytes_per_sector', got '$divider'" >&2
  exit 1
fi

# for each passed-in dmesg block number do
while test "$#" -gt "0"; do
dmesgblock="$1"
shift
remaining_args=("$@") #for on_exit() above
echo '--------'
echo "Passed-in dmesg block($bytes_per_sector bytes per block) number: '$dmesgblock'"
#have to handle the case when $dmesgblock is empty and when it's negative eg. "-1" so using a default value(of 0) when unset in the below 'if' block will help not break the 'test' expecting an integer while also allowing negative numbers ("0$dmesgblock" would otherwise yield "0-1" a non-integer):
if test "${dmesgblock:-0}" -le "0"; then
  echo "Bad passed-in dmesg block number: '$dmesgblock'" >&2
  exit 1
fi

#TODO: check and fail(or warn? nah, it should be fail!) if not a multiplier (eg. use modullo? is it "%" ?)
block=$(( $dmesgblock / 8 ))
if ! test "${block:--1}" -ge "0"; then
  echo "Badly computed device block number: '$block'" >&2
  exit 1
fi

echo "Actual block number(of $blocksize bytes per block): $block"
inode="$(echo "open ${dev}"$'\n'"icheck ${block}"$'\n'"close" | $sudo debugfs -f - 2>/dev/null | tail -n2|head -1|cut -f2 -d$'\t')"
if test "<block not found>" == "$inode"; then
  echo "No inode was found for the provided dmesg block number '$dmesgblock' which mapped to dev block number '$block'" >&2
  exit 1
else
    #assuming number TODO: check for this!
    echo "Found inode: $inode"
    fpath="$(echo "open ${dev}"$'\n'"ncheck ${inode}"$'\n'"close" | $sudo debugfs -f - 2>/dev/null | tail -n2|head -1|cut -f2- -d$'\t')"
  #fpath always begins with '/' right?
    if test "$fpath" != "Pathname"; then
        echo "Found path : $fpath"
    else
        echo "not found"
    fi
fi
done

示例用法:

[user@dev01-w-s-f-fdr28 ~]$ ./showblock  /dev/xvda3 5379184 5129952 "" -1 "" 1200 ; echo "exit code: $?"
--------
Passed-in dmesg block(512 bytes per block) number: '5379184'
Actual block number(of 4096 bytes per block): 672398
Found inode: 134571
Found path : /usr/bin/sleep
--------
Passed-in dmesg block(512 bytes per block) number: '5129952'
Actual block number(of 4096 bytes per block): 641244
Found inode: 141497
Found path : /usr/lib64/libIlmImf-2_2.so.22.0.0
--------
Passed-in dmesg block(512 bytes per block) number: ''
Bad passed-in dmesg block number: ''
WARNING: There are '3' remaining args not processed, they are: '-1' '' '1200' 
exit code: 1
[user@dev01-w-s-f-fdr28 ~]$ ./showblock  /dev/xvda3 5379184 5129952 -1 "" "" 1200 ; echo "exit code: $?"
--------
Passed-in dmesg block(512 bytes per block) number: '5379184'
Actual block number(of 4096 bytes per block): 672398
Found inode: 134571
Found path : /usr/bin/sleep
--------
Passed-in dmesg block(512 bytes per block) number: '5129952'
Actual block number(of 4096 bytes per block): 641244
Found inode: 141497
Found path : /usr/lib64/libIlmImf-2_2.so.22.0.0
--------
Passed-in dmesg block(512 bytes per block) number: '-1'
Bad passed-in dmesg block number: '-1'
WARNING: There are '3' remaining args not processed, they are: '' '' '1200' 
exit code: 1
[user@dev01-w-s-f-fdr28 ~]$ ./showblock  /dev/xvda3 5379184 5129952 0 ; echo "exit code: $?"
--------
Passed-in dmesg block(512 bytes per block) number: '5379184'
Actual block number(of 4096 bytes per block): 672398
Found inode: 134571
Found path : /usr/bin/sleep
--------
Passed-in dmesg block(512 bytes per block) number: '5129952'
Actual block number(of 4096 bytes per block): 641244
Found inode: 141497
Found path : /usr/lib64/libIlmImf-2_2.so.22.0.0
--------
Passed-in dmesg block(512 bytes per block) number: '0'
Bad passed-in dmesg block number: '0'
exit code: 1
[user@dev01-w-s-f-fdr28 ~]$ ./showblock  /dev/xvda3 5379184 5129952 3 ; echo "exit code: $?"
--------
Passed-in dmesg block(512 bytes per block) number: '5379184'
Actual block number(of 4096 bytes per block): 672398
Found inode: 134571
Found path : /usr/bin/sleep
--------
Passed-in dmesg block(512 bytes per block) number: '5129952'
Actual block number(of 4096 bytes per block): 641244
Found inode: 141497
Found path : /usr/lib64/libIlmImf-2_2.so.22.0.0
--------
Passed-in dmesg block(512 bytes per block) number: '3'
Actual block number(of 4096 bytes per block): 0
No inode was found for the provided dmesg block number '3' which mapped to dev block number '0'
exit code: 1
[user@dev01-w-s-f-fdr28 ~]$ ./showblock  /dev/xvda3 5379184 5129952 ; echo "exit code: $?"
--------
Passed-in dmesg block(512 bytes per block) number: '5379184'
Actual block number(of 4096 bytes per block): 672398
Found inode: 134571
Found path : /usr/bin/sleep
--------
Passed-in dmesg block(512 bytes per block) number: '5129952'
Actual block number(of 4096 bytes per block): 641244
Found inode: 141497
Found path : /usr/lib64/libIlmImf-2_2.so.22.0.0
exit code: 0

这是一个处理整个 dmesg 日志的脚本:

#!/bin/bash

#./showallblocks rev.01 rewritten for question/answer from: https://stackoverflow.com/q/52058914/10239615

if test "`id -u`" != "0"; then
    sudo='sudo'
else
    sudo=''
fi

dmesglog="$1"
if test -z "$dmesglog"; then
  echo "Usage: '$0' <dmesglogfile>"
  echo "Examples:"
  echo "sudo dmesg > dmesg1.log && '$0' dmesg1.log"
  echo "'$0' <(sudo dmesg)"
  #Note: '$0' used for the case when $0 has spaces or other things in its path names, and user wants to copy paste, for whatever reason, the output of the above into the command line.
  exit 1
fi

#(optional) Stop logging if already in progress:
$sudo sysctl -w vm.block_dump=0

#Using the answer from here(thanks to glenn jackman): https://unix.stackexchange.com/a/467377/306023
#grep --color=never -E -- 'READ block [0-9]+ on xvda3' "$dmesglog" |
#cat "$dmesglog" |
$sudo perl -pe '
if (! /READ block [0-9]+ on [A-Za-z0-9]+ .*$/) {
  s{.*}{}s
}

s{(READ block) (\d+) (on) ([A-Za-z0-9]+) ([^\$]*)\n$}
{join " ",$1, $2, $3, $4, $5, qx(./showblock "/dev/$4" "$2" | grep -F -- "Found path :" | cut -f4- -d" ")}es
' -- "$dmesglog"
#Note: the output of qx(...) above is purposely allowed to have trailing newline! (I did wonder if purposely is correct here or it should be purposefully, https://www.merriam-webster.com/words-at-play/purposely-purposefully-usage )
#To find out what "}es"(above) is, see perlre modifiers: https://perldoc.perl.org/perlre.html#Modifiers
#FIXME: noobish try to exclude lines not matching the lines that need to be replaced, from output! used 'if' above

样本输出:

[user@dev01-w-s-f-fdr28 ~]$ ./showallblocks sample.dmesg.log 
vm.block_dump = 0
[ 6031.953619] sysctl(20774): READ block 5285528 on xvda3 (32 sectors) /usr/lib64/libgpg-error.so.0.24.2
[ 6031.954317] sysctl(20774): READ block 5285768 on xvda3 (8 sectors) /usr/lib64/libgpg-error.so.0.24.2
[ 6031.954598] sysctl(20774): READ block 5285648 on xvda3 (120 sectors) /usr/lib64/libgpg-error.so.0.24.2
[ 6031.954617] sysctl(20774): READ block 5285776 on xvda3 (24 sectors) /usr/lib64/libgpg-error.so.0.24.2
[ 6031.955482] sysctl(20774): READ block 5285560 on xvda3 (88 sectors) /usr/lib64/libgpg-error.so.0.24.2
[ 6031.955699] sysctl(20774): READ block 4473568 on xvda3 (8 sectors) /usr/lib64/libuuid.so.1.3.0
[ 6031.955730] sysctl(20774): READ block 4473584 on xvda3 (16 sectors) /usr/lib64/libuuid.so.1.3.0
[ 6031.955787] sysctl(20774): READ block 4749496 on xvda3 (224 sectors) /usr/lib64/libblkid.so.1.1.0
[ 6030.638492] bash(20757): READ block 4543096 on xvda3 (32 sectors) /usr/bin/gawk
[ 6030.639133] awk(20757): READ block 4544280 on xvda3 (176 sectors) /usr/bin/gawk
于 2018-09-07T22:07:07.210 回答