kexec-tools/mkdumprd
Kairui Song 539bff4083 Append both nofail and x-systemd.before to kdump mount target
By this point, there is still an unresolved vfs kernel issue that blocks
systemd from mounting the dump target properly from time to time. To
prevent systemd from failing by mounting the dump target, we can add
nofail option to the kdump mount point.

But adding nofail will wipe out default dependency of the mount point,
see commit 94a7b43, so systemd randomize the order of calling kdump.sh
and mounting the dump target and lead to unexpected behavior.
However we can use x-systemd.before to ensure the mount is done
in right order.

In dracut-kdump-capture.service, we have "After=initrd.target",
and look at dracut.bootup.7, systems start processing of fstab by
initrd-fs.target, so set "After=initrd.target" could ensure the mount
is ready before pre-pivot and kdump-capture service.

Now with both nofail and x-systemd.before=initrd.target, systemd
will try to mount the dump target before calling kdump, and even if the
mount failed, kdump.sh will still be called and try to mount again. See
dump_fs function, which will try to mount if the target is not mounted.
Kdump will only fail if both mount attemp fails.

Else if the kdump target mount failed or unstable, systemd will directly
jump to kdump failure action, and kdump fails.

This should improve the robustness in general with no other risk.

Signed-off-by: Kairui Song <kasong@redhat.com>
Acked-by: Pingfan Liu <piliu@redhat.com>
2020-05-15 17:17:27 +08:00

442 lines
11 KiB
Bash

#!/bin/bash --norc
# New mkdumprd
#
# Copyright 2011 Red Hat, Inc.
#
# Written by Cong Wang <amwang@redhat.com>
#
[[ $dracutbasedir ]] || dracutbasedir=/usr/lib/dracut
. $dracutbasedir/dracut-functions.sh
. /lib/kdump/kdump-lib.sh
export IN_KDUMP=1
conf_file="/etc/kdump.conf"
SSH_KEY_LOCATION="/root/.ssh/kdump_id_rsa"
SAVE_PATH=$(get_save_path)
OVERRIDE_RESETTABLE=0
extra_modules=""
dracut_args="--quiet --hostonly --hostonly-cmdline --hostonly-i18n --hostonly-mode strict -o \"plymouth dash resume ifcfg earlykdump\""
is_wdt_addition_needed() {
local active
is_wdt_mod_omitted
[[ $? -eq 0 ]] && return 1
[[ -d /sys/class/watchdog/ ]] || return 1
for dir in /sys/class/watchdog/*; do
[[ -f "$dir/state" ]] || continue
active=$(< "$dir/state")
[[ "$active" = "active" ]] && return 0
done
return 1
}
add_dracut_arg() {
dracut_args="$dracut_args $@"
}
add_dracut_module() {
add_dracut_arg "--add" "\"$1\""
}
add_dracut_mount() {
add_dracut_arg "--mount" "\"$1\""
}
add_dracut_sshkey() {
add_dracut_arg "--sshkey" "\"$1\""
}
# caller should ensure $1 is valid and mounted in 1st kernel
to_mount() {
local _dev=$1 _source _target _fstype _options _mntopts _pdev
_source=$(findmnt -k -f -n -r -o SOURCE $_dev)
_target=$(get_mntpoint_from_target $_dev)
# mount under /sysroot if dump to root disk or mount under
#/kdumproot/$_target in other cases in 2nd kernel. systemd
#will be in charge to umount it.
if [ "$_target" = "/" ];then
_target="/sysroot"
else
_target="/kdumproot/$_target"
fi
_fstype=$(findmnt -k -f -n -r -o FSTYPE $_dev)
[[ -e /etc/fstab ]] && _options=$(findmnt --fstab -f -n -r -o OPTIONS $_dev)
if [ -z "$_options" ]; then
_options=$(findmnt -k -f -n -r -o OPTIONS $_dev)
if [[ $_fstype == "nfs"* ]]; then
_options=$(echo $_options | sed 's/,addr=[^,]*//')
_options=$(echo $_options | sed 's/,proto=[^,]*//')
_options=$(echo $_options | sed 's/,clientaddr=[^,]*//')
fi
fi
# mount fs target as rw in 2nd kernel
_options=$(echo $_options | sed 's/\(^\|,\)ro\($\|,\)/\1rw\2/g')
# with 'noauto' in fstab nfs and non-root disk mount will fail in 2nd
# kernel, filter it out here.
_options=$(echo $_options | sed 's/\(^\|,\)noauto\($\|,\)/\1/g')
# drop nofail or nobootwait
_options=$(echo $_options | sed 's/\(^\|,\)nofail\($\|,\)/\1/g')
_options=$(echo $_options | sed 's/\(^\|,\)nobootwait\($\|,\)/\1/g')
# use both nofail and x-systemd.before to ensure systemd will try best to
# mount it before kdump starts, this is an attempt to improve robustness
_options="$_options,nofail,x-systemd.before=initrd-fs.target"
_mntopts="$_target $_fstype $_options"
#for non-nfs _dev converting to use udev persistent name
if [ -b "$_source" ]; then
_pdev="$(get_persistent_dev $_source)"
if [ -z "$_pdev" ]; then
return 1
fi
else
_pdev=$_dev
fi
echo "$_pdev $_mntopts"
}
is_readonly_mount() {
local _mnt
_mnt=$(findmnt -k -f -n -r -o OPTIONS $1)
#fs/proc_namespace.c: show_mountinfo():
#seq_puts(m, mnt->mnt_flags & MNT_READONLY ? " ro" : " rw");
[[ "$_mnt" =~ ^ro ]]
}
#Function: get_ssh_size
#$1=dump target
#called from while loop and shouldn't read from stdin, so we're using "ssh -n"
get_ssh_size() {
local _opt _out _size
_opt="-i $SSH_KEY_LOCATION -o BatchMode=yes -o StrictHostKeyChecking=yes"
_out=$(ssh -q -n $_opt $1 "df -P $SAVE_PATH")
[ $? -ne 0 ] && {
perror_exit "checking remote ssh server available size failed."
}
#ssh output removed the line break, so print field NF-2
_size=$(echo -n $_out| awk '{avail=NF-2; print $avail}')
echo -n $_size
}
#mkdir if save path does not exist on ssh dump target
#$1=ssh dump target
#caller should ensure write permission on $1:$SAVE_PATH
#called from while loop and shouldn't read from stdin, so we're using "ssh -n"
mkdir_save_path_ssh()
{
local _opt _dir
_opt="-i $SSH_KEY_LOCATION -o BatchMode=yes -o StrictHostKeyChecking=yes"
ssh -qn $_opt $1 mkdir -p $SAVE_PATH 2>&1 > /dev/null
_ret=$?
if [ $_ret -ne 0 ]; then
perror_exit "mkdir failed on $1:$SAVE_PATH"
fi
#check whether user has write permission on $1:$SAVE_PATH
_dir=$(ssh -qn $_opt $1 mktemp -dqp $SAVE_PATH 2>/dev/null)
_ret=$?
if [ $_ret -ne 0 ]; then
perror_exit "Could not create temporary directory on $1:$SAVE_PATH. Make sure user has write permission on destination"
fi
ssh -qn $_opt $1 rmdir $_dir
return 0
}
#Function: get_fs_size
#$1=dump target
get_fs_size() {
local _mnt=$(get_mntpoint_from_target $1)
echo -n $(df -P "${_mnt}/$SAVE_PATH"|tail -1|awk '{print $4}')
}
#Function: get_raw_size
#$1=dump target
get_raw_size() {
echo -n $(fdisk -s "$1")
}
#Function: check_size
#$1: dump type string ('raw', 'fs', 'ssh')
#$2: dump target
check_size() {
local avail memtotal
memtotal=$(awk '/MemTotal/{print $2}' /proc/meminfo)
case "$1" in
raw)
avail=$(get_raw_size "$2")
;;
ssh)
avail=$(get_ssh_size "$2")
;;
fs)
avail=$(get_fs_size "$2")
;;
*)
return
esac
if [ $? -ne 0 ]; then
perror_exit "Check dump target size failed"
fi
if [ $avail -lt $memtotal ]; then
echo "Warning: There might not be enough space to save a vmcore."
echo " The size of $2 should be greater than $memtotal kilo bytes."
fi
}
# $1: core_collector config value
verify_core_collector() {
local _cmd="${1%% *}"
local _params="${1#* }"
if [ "$_cmd" != "makedumpfile" ]; then
if is_raw_dump_target; then
echo "Warning: specifying a non-makedumpfile core collector, you will have to recover the vmcore manually."
fi
return
fi
if is_ssh_dump_target || is_raw_dump_target; then
if ! strstr "$_params" "-F"; then
perror_exit "The specified dump target needs makedumpfile \"-F\" option."
fi
_params="$_params vmcore"
else
_params="$_params vmcore dumpfile"
fi
if ! $_cmd --check-params $_params; then
perror_exit "makedumpfile parameter check failed."
fi
}
add_mount() {
local _mnt=$(to_mount "$1")
if [ $? -ne 0 ]; then
exit 1
fi
add_dracut_mount "$_mnt"
}
#handle the case user does not specify the dump target explicitly
handle_default_dump_target()
{
local _target
local _mntpoint
is_user_configured_dump_target && return
check_save_path_fs $SAVE_PATH
_save_path=$(get_bind_mount_source $SAVE_PATH)
_target=$(get_target_from_path $_save_path)
_mntpoint=$(get_mntpoint_from_target $_target)
SAVE_PATH=${_save_path##"$_mntpoint"}
add_mount "$_target"
check_size fs $_target
}
get_override_resettable()
{
local override_resettable
override_resettable=$(grep "^override_resettable" $conf_file)
if [ -n "$override_resettable" ]; then
OVERRIDE_RESETTABLE=$(echo $override_resettable | cut -d' ' -f2)
if [ "$OVERRIDE_RESETTABLE" != "0" ] && [ "$OVERRIDE_RESETTABLE" != "1" ];then
perror_exit "override_resettable value $OVERRIDE_RESETTABLE is invalid"
fi
fi
}
# $1: function name
for_each_block_target()
{
local dev majmin
for dev in $(get_kdump_targets); do
[ -b "$dev" ] || continue
majmin=$(get_maj_min $dev)
check_block_and_slaves $1 $majmin && return 1
done
return 0
}
#judge if a specific device with $1 is unresettable
#return false if unresettable.
is_unresettable()
{
local path="/sys/$(udevadm info --query=all --path=/sys/dev/block/$1 | awk '/^P:/ {print $2}' | sed -e 's/\(cciss[0-9]\+\/\).*/\1/g' -e 's/\/block\/.*$//')/resettable"
local resettable=1
if [ -f "$path" ]
then
resettable="$(cat $path)"
[ $resettable -eq 0 -a "$OVERRIDE_RESETTABLE" -eq 0 ] && {
local device=$(udevadm info --query=all --path=/sys/dev/block/$1 | awk -F= '/DEVNAME/{print $2}')
echo "Error: Can not save vmcore because device $device is unresettable"
return 0
}
fi
return 1
}
#check if machine is resettable.
#return true if resettable
check_resettable()
{
local _ret _target
get_override_resettable
for_each_block_target is_unresettable
_ret=$?
[ $_ret -eq 0 ] && return
return 1
}
# $1: maj:min
is_crypt()
{
local majmin=$1 dev line ID_FS_TYPE=""
line=$(udevadm info --query=property --path=/sys/dev/block/$majmin \
| grep "^ID_FS_TYPE")
eval "$line"
[[ "$ID_FS_TYPE" = "crypto_LUKS" ]] && {
dev=$(udevadm info --query=all --path=/sys/dev/block/$majmin | awk -F= '/DEVNAME/{print $2}')
echo "Device $dev is encrypted."
return 0
}
return 1
}
check_crypt()
{
local _ret _target
for_each_block_target is_crypt
_ret=$?
[ $_ret -eq 0 ] && return
return 1
}
if ! check_resettable; then
exit 1
fi
if ! check_crypt; then
echo "Warning: Encrypted device is in dump path. User will prompted for password during second kernel boot."
fi
# firstly get right SSH_KEY_LOCATION
keyfile=$(awk '/^sshkey/ {print $2}' $conf_file)
if [ -f "$keyfile" ]; then
# canonicalize the path
SSH_KEY_LOCATION=$(/usr/bin/readlink -m $keyfile)
fi
if [ "$(uname -m)" = "s390x" ]; then
add_dracut_module "znet"
fi
if is_wdt_addition_needed; then
add_dracut_arg "-a" "watchdog"
fi
while read config_opt config_val;
do
# remove inline comments after the end of a directive.
case "$config_opt" in
extra_modules)
extra_modules="$extra_modules $config_val"
;;
ext[234]|xfs|btrfs|minix|nfs)
if ! findmnt $config_val >/dev/null; then
perror_exit "Dump target $config_val is probably not mounted."
fi
# User configured target, use $SAVE_PATH as the dump path within the target
check_save_path_user_configured "$config_val" "$SAVE_PATH"
check_size fs "$config_val"
add_mount "$config_val"
;;
raw)
# checking raw disk writable
dd if=$config_val count=1 of=/dev/null > /dev/null 2>&1 || {
perror_exit "Bad raw disk $config_val"
}
_praw=$(persistent_policy="by-id" get_persistent_dev $config_val)
if [ -z "$_praw" ]; then
exit 1
fi
add_dracut_arg "--device" "$_praw"
check_size raw $config_val
;;
ssh)
if strstr "$config_val" "@";
then
mkdir_save_path_ssh $config_val
check_size ssh $config_val
add_dracut_module "ssh-client"
add_dracut_sshkey "$SSH_KEY_LOCATION"
else
perror_exit "Bad ssh dump target $config_val"
fi
;;
core_collector)
verify_core_collector "$config_val"
;;
dracut_args)
add_dracut_arg $config_val
;;
*)
;;
esac
done <<< "$(read_strip_comments $conf_file)"
handle_default_dump_target
if [ -n "$extra_modules" ]
then
add_dracut_arg "--add-drivers" "$extra_modules"
fi
if ! is_fadump_capable; then
# The 2nd rootfs mount stays behind the normal dump target mount,
# so it doesn't affect the logic of check_dump_fs_modified().
is_dump_to_rootfs && add_mount "$(to_dev_name $(get_root_fs_device))"
add_dracut_arg "--no-hostonly-default-device"
fi
echo "$dracut_args $@" | xargs dracut
_rc=$?
sync
exit $_rc