2017-05-09 09:30:30 +00:00
|
|
|
#!/bin/bash
|
2011-07-06 19:25:34 +00:00
|
|
|
KEXEC=/sbin/kexec
|
|
|
|
|
|
|
|
KDUMP_KERNELVER=""
|
|
|
|
KDUMP_COMMANDLINE=""
|
|
|
|
KEXEC_ARGS=""
|
kdump-lib: switch to the kexec_file_load() syscall on x86_64 by default
UEFI Secure boot is a signature verification mechanism, designed to
prevent malicious code being loaded and executed at the early boot
stage. This makes sure that code executed is trusted by firmware.
Previously, with kexec_file_load() interface, kernel prevents unsigned
kernel image from being loaded if secure boot is enabled. So kdump will
detect whether secure boot is enabled firstly, then decide which interface
is chosen to execute, kexec_load() or kexec_file_load(). Otherwise unsigned
kernel loading will fail if secure boot enabled, and kexec_file_load() is
entered.
Now, the implementation of kexec_file_load() is adjusted in below commit.
With this change, if CONFIG_KEXEC_SIG_FORCE is not set, unsigned kernel
still has a chance to be allowed to load under some conditions.
commit 99d5cadfde2b ("kexec_file: split KEXEC_VERIFY_SIG into KEXEC_SIG
and KEXEC_SIG_FORCE")
And in the current Fedora, the CONFIG_KEXEC_SIG_FORCE is not set, only the
CONFIG_KEXEC_SIG and CONFIG_BZIMAGE_VERIFY_SIG are set on x86_64 by default.
It's time to spread kexec_file_load() onto all systems of x86_64, including
Secure-boot platforms and legacy platforms. Please refer to the following
form.
.----------------------------------------------------------------------.
| . | signed kernel | unsigned kernel |
| . types |-----------------------|-----------------------|
| . |Secure boot| Legacy |Secure boot| Legacy |
| . |-----------|-----------|-----------|-----------|
| options . | prev| now | prev| now | | | prev| now |
| . |(file|(file|(only|(file| prev| now |(only|(file|
| . |load)|load)|load)|load)| | |load)|load)|
|----------------------|-----|-----|-----|-----|-----|-----|-----|-----|
|KEXEC_SIG=y | | | | | | | | |
|SIG_FORCE is not set |succ |succ |succ |succ | X | X |succ |succ |
|BZIMAGE_VERIFY_SIG=y | | | | | | | | |
|----------------------|-----|-----|-----|-----|-----|-----|-----|-----|
|KEXEC_SIG=y | | | | | | | | |
|SIG_FORCE is not set | | | | | | | | |
|BZIMAGE_VERIFY_SIG is |fail |fail |succ |fail | X | X |succ |fail |
|not set | | | | | | | | |
|----------------------|-----|-----|-----|-----|-----|-----|-----|-----|
|KEXEC_SIG=y | | | | | | | | |
|SIG_FORCE=y |succ |succ |succ |fail | X | X |succ |fail |
|BZIMAGE_VERIFY_SIG=y | | | | | | | | |
|----------------------|-----|-----|-----|-----|-----|-----|-----|-----|
|KEXEC_SIG=y | | | | | | | | |
|SIG_FORCE=y | | | | | | | | |
|BZIMAGE_VERIFY_SIG is |fail |fail |succ |fail | X | X |succ |fail |
|not set | | | | | | | | |
|----------------------|-----|-----|-----|-----|-----|-----|-----|-----|
|KEXEC_SIG is not set | | | | | | | | |
|SIG_FORCE is not set | | | | | | | | |
|BZIMAGE_VERIFY_SIG is |fail |fail |succ |succ | X | X |succ |succ |
|not set | | | | | | | | |
----------------------------------------------------------------------
Note:
[1] The 'X' indicates that the 1st kernel(unsigned) can not boot when the
Secure boot is enabled.
Hence, in this patch, if on x86_64, let's use the kexec_file_load() only.
See if anything wrong happened in this case, in Fedora firstly for the
time being.
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
Acked-by: Kairui Song <kasong@redhat.com>
2020-01-16 05:47:35 +00:00
|
|
|
KDUMP_FILE_LOAD=""
|
2011-07-06 19:25:34 +00:00
|
|
|
KDUMP_CONFIG_FILE="/etc/kdump.conf"
|
2012-06-06 08:24:23 +00:00
|
|
|
MKDUMPRD="/sbin/mkdumprd -f"
|
2017-08-31 06:27:00 +00:00
|
|
|
DRACUT_MODULES_FILE="/usr/lib/dracut/modules.txt"
|
2012-02-22 03:16:09 +00:00
|
|
|
SAVE_PATH=/var/crash
|
|
|
|
SSH_KEY_LOCATION="/root/.ssh/kdump_id_rsa"
|
2016-11-03 18:46:31 +00:00
|
|
|
INITRD_CHECKSUM_LOCATION="/boot/.fadump_initrd_checksum"
|
2012-02-22 03:16:09 +00:00
|
|
|
DUMP_TARGET=""
|
2016-11-03 18:46:31 +00:00
|
|
|
DEFAULT_INITRD=""
|
|
|
|
DEFAULT_INITRD_BAK=""
|
kdump: Rebuild default initrd for firmware assisted dump
The current kdump infrastructure builds a separate initrd which then
gets loaded into memory by kexec-tools for use by kdump kernel. But
firmware assisted dump (FADUMP) does not use kexec-based approach.
After crash, firmware reboots the partition and loads grub loader
like the normal booting process does. Hence in the FADUMP approach,
the second kernel (after crash) will always use the default initrd
(OS built). So, to support FADUMP, change is required, as in to add
dump capturing steps, in this initrd.
The current kdumpctl script implementation already has the code to
build initrd using mkdumprd. This patch uses the new '--rebuild'
option introduced, in dracut, to incrementally build the initramfs
image. Before rebuilding, we may need to probe the initrd image for
fadump support, to avoid rebuilding the initrd image multiple times
unnecessarily. This can be done using "lsinitrd" tool with the newly
proposed '--mod' option & inspecting the presence of "kdumpbase" in
the list of modules of default initrd image. We rebuild the image if
only "kdumpbase" module is missing in the initrd image. Also, before
rebuilding, a backup of default initrd image is taken.
Kexec-tools package in rhel7 is now enhanced to insert a out-of-tree
kdump module for dracut, which is responsible for adding vmcore
capture steps into initrd, if dracut is invoked with "IN_KDUMP"
environment variable set to 1. mkdumprd script exports "IN_KDUMP=1"
environment variable before invoking dracut to build kdump initrd.
This patch relies on this current mechanism of kdump init script.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
2014-07-24 18:39:07 +00:00
|
|
|
TARGET_INITRD=""
|
2014-07-24 18:38:36 +00:00
|
|
|
FADUMP_REGISTER_SYS_NODE="/sys/kernel/fadump_registered"
|
|
|
|
#kdump shall be the default dump mode
|
|
|
|
DEFAULT_DUMP_MODE="kdump"
|
2016-05-09 08:17:53 +00:00
|
|
|
image_time=0
|
2011-07-06 19:25:34 +00:00
|
|
|
|
2016-11-17 04:51:12 +00:00
|
|
|
[[ $dracutbasedir ]] || dracutbasedir=/usr/lib/dracut
|
|
|
|
. $dracutbasedir/dracut-functions.sh
|
2013-09-26 11:46:48 +00:00
|
|
|
. /lib/kdump/kdump-lib.sh
|
|
|
|
|
2011-07-06 19:25:34 +00:00
|
|
|
standard_kexec_args="-p"
|
|
|
|
|
2015-11-18 08:00:22 +00:00
|
|
|
# Some default values in case /etc/sysconfig/kdump doesn't include
|
|
|
|
KDUMP_COMMANDLINE_REMOVE="hugepages hugepagesz slub_debug"
|
|
|
|
|
2011-07-06 19:25:34 +00:00
|
|
|
if [ -f /etc/sysconfig/kdump ]; then
|
|
|
|
. /etc/sysconfig/kdump
|
|
|
|
fi
|
|
|
|
|
2013-08-30 09:21:56 +00:00
|
|
|
single_instance_lock()
|
|
|
|
{
|
2014-07-16 12:09:48 +00:00
|
|
|
local rc timeout=5
|
|
|
|
|
2013-08-30 09:21:56 +00:00
|
|
|
exec 9>/var/lock/kdump
|
2017-07-12 02:59:55 +00:00
|
|
|
if [ $? -ne 0 ]; then
|
|
|
|
echo "Create file lock failed"
|
|
|
|
exit 1
|
|
|
|
fi
|
2014-07-16 12:09:48 +00:00
|
|
|
|
|
|
|
flock -n 9
|
|
|
|
rc=$?
|
|
|
|
|
|
|
|
while [ $rc -ne 0 ]; do
|
|
|
|
echo "Another app is currently holding the kdump lock; waiting for it to exit..."
|
|
|
|
flock -w $timeout 9
|
|
|
|
rc=$?
|
|
|
|
done
|
2013-08-30 09:21:56 +00:00
|
|
|
}
|
|
|
|
|
2014-07-24 18:38:36 +00:00
|
|
|
determine_dump_mode()
|
|
|
|
{
|
|
|
|
# Check if firmware-assisted dump is enabled
|
|
|
|
# if yes, set the dump mode as fadump
|
|
|
|
if is_fadump_capable; then
|
|
|
|
echo "Dump mode is fadump"
|
|
|
|
DEFAULT_DUMP_MODE="fadump"
|
|
|
|
fi
|
|
|
|
}
|
|
|
|
|
2014-07-15 06:41:54 +00:00
|
|
|
save_core()
|
2011-07-06 19:25:34 +00:00
|
|
|
{
|
|
|
|
coredir="/var/crash/`date +"%Y-%m-%d-%H:%M"`"
|
|
|
|
|
|
|
|
mkdir -p $coredir
|
|
|
|
cp --sparse=always /proc/vmcore $coredir/vmcore-incomplete
|
|
|
|
if [ $? == 0 ]; then
|
|
|
|
mv $coredir/vmcore-incomplete $coredir/vmcore
|
2013-02-18 09:05:41 +00:00
|
|
|
echo "saved a vmcore to $coredir"
|
2011-07-06 19:25:34 +00:00
|
|
|
else
|
2013-02-18 09:05:41 +00:00
|
|
|
echo "failed to save a vmcore to $coredir" >&2
|
2011-07-06 19:25:34 +00:00
|
|
|
fi
|
|
|
|
|
|
|
|
# pass the dmesg to Abrt tool if exists, in order
|
|
|
|
# to collect the kernel oops message.
|
|
|
|
# https://fedorahosted.org/abrt/
|
|
|
|
if [ -x /usr/bin/dumpoops ]; then
|
|
|
|
makedumpfile --dump-dmesg $coredir/vmcore $coredir/dmesg >/dev/null 2>&1
|
|
|
|
dumpoops -d $coredir/dmesg >/dev/null 2>&1
|
|
|
|
if [ $? == 0 ]; then
|
2013-02-18 09:05:41 +00:00
|
|
|
echo "kernel oops has been collected by abrt tool"
|
2011-07-06 19:25:34 +00:00
|
|
|
fi
|
|
|
|
fi
|
|
|
|
}
|
|
|
|
|
kdump: Rebuild default initrd for firmware assisted dump
The current kdump infrastructure builds a separate initrd which then
gets loaded into memory by kexec-tools for use by kdump kernel. But
firmware assisted dump (FADUMP) does not use kexec-based approach.
After crash, firmware reboots the partition and loads grub loader
like the normal booting process does. Hence in the FADUMP approach,
the second kernel (after crash) will always use the default initrd
(OS built). So, to support FADUMP, change is required, as in to add
dump capturing steps, in this initrd.
The current kdumpctl script implementation already has the code to
build initrd using mkdumprd. This patch uses the new '--rebuild'
option introduced, in dracut, to incrementally build the initramfs
image. Before rebuilding, we may need to probe the initrd image for
fadump support, to avoid rebuilding the initrd image multiple times
unnecessarily. This can be done using "lsinitrd" tool with the newly
proposed '--mod' option & inspecting the presence of "kdumpbase" in
the list of modules of default initrd image. We rebuild the image if
only "kdumpbase" module is missing in the initrd image. Also, before
rebuilding, a backup of default initrd image is taken.
Kexec-tools package in rhel7 is now enhanced to insert a out-of-tree
kdump module for dracut, which is responsible for adding vmcore
capture steps into initrd, if dracut is invoked with "IN_KDUMP"
environment variable set to 1. mkdumprd script exports "IN_KDUMP=1"
environment variable before invoking dracut to build kdump initrd.
This patch relies on this current mechanism of kdump init script.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
2014-07-24 18:39:07 +00:00
|
|
|
rebuild_fadump_initrd()
|
2012-06-14 01:57:27 +00:00
|
|
|
{
|
kdump: Rebuild default initrd for firmware assisted dump
The current kdump infrastructure builds a separate initrd which then
gets loaded into memory by kexec-tools for use by kdump kernel. But
firmware assisted dump (FADUMP) does not use kexec-based approach.
After crash, firmware reboots the partition and loads grub loader
like the normal booting process does. Hence in the FADUMP approach,
the second kernel (after crash) will always use the default initrd
(OS built). So, to support FADUMP, change is required, as in to add
dump capturing steps, in this initrd.
The current kdumpctl script implementation already has the code to
build initrd using mkdumprd. This patch uses the new '--rebuild'
option introduced, in dracut, to incrementally build the initramfs
image. Before rebuilding, we may need to probe the initrd image for
fadump support, to avoid rebuilding the initrd image multiple times
unnecessarily. This can be done using "lsinitrd" tool with the newly
proposed '--mod' option & inspecting the presence of "kdumpbase" in
the list of modules of default initrd image. We rebuild the image if
only "kdumpbase" module is missing in the initrd image. Also, before
rebuilding, a backup of default initrd image is taken.
Kexec-tools package in rhel7 is now enhanced to insert a out-of-tree
kdump module for dracut, which is responsible for adding vmcore
capture steps into initrd, if dracut is invoked with "IN_KDUMP"
environment variable set to 1. mkdumprd script exports "IN_KDUMP=1"
environment variable before invoking dracut to build kdump initrd.
This patch relies on this current mechanism of kdump init script.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
2014-07-24 18:39:07 +00:00
|
|
|
local target_initrd_tmp
|
|
|
|
|
2014-07-24 18:39:22 +00:00
|
|
|
# this file tells the initrd is fadump enabled
|
|
|
|
touch /tmp/fadump.initramfs
|
kdump: Rebuild default initrd for firmware assisted dump
The current kdump infrastructure builds a separate initrd which then
gets loaded into memory by kexec-tools for use by kdump kernel. But
firmware assisted dump (FADUMP) does not use kexec-based approach.
After crash, firmware reboots the partition and loads grub loader
like the normal booting process does. Hence in the FADUMP approach,
the second kernel (after crash) will always use the default initrd
(OS built). So, to support FADUMP, change is required, as in to add
dump capturing steps, in this initrd.
The current kdumpctl script implementation already has the code to
build initrd using mkdumprd. This patch uses the new '--rebuild'
option introduced, in dracut, to incrementally build the initramfs
image. Before rebuilding, we may need to probe the initrd image for
fadump support, to avoid rebuilding the initrd image multiple times
unnecessarily. This can be done using "lsinitrd" tool with the newly
proposed '--mod' option & inspecting the presence of "kdumpbase" in
the list of modules of default initrd image. We rebuild the image if
only "kdumpbase" module is missing in the initrd image. Also, before
rebuilding, a backup of default initrd image is taken.
Kexec-tools package in rhel7 is now enhanced to insert a out-of-tree
kdump module for dracut, which is responsible for adding vmcore
capture steps into initrd, if dracut is invoked with "IN_KDUMP"
environment variable set to 1. mkdumprd script exports "IN_KDUMP=1"
environment variable before invoking dracut to build kdump initrd.
This patch relies on this current mechanism of kdump init script.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
2014-07-24 18:39:07 +00:00
|
|
|
target_initrd_tmp="$TARGET_INITRD.tmp"
|
2019-02-05 20:17:38 +00:00
|
|
|
$MKDUMPRD $target_initrd_tmp --rebuild $DEFAULT_INITRD_BAK --kver $kdump_kver \
|
2014-07-24 18:39:22 +00:00
|
|
|
-i /tmp/fadump.initramfs /etc/fadump.initramfs
|
kdump: Rebuild default initrd for firmware assisted dump
The current kdump infrastructure builds a separate initrd which then
gets loaded into memory by kexec-tools for use by kdump kernel. But
firmware assisted dump (FADUMP) does not use kexec-based approach.
After crash, firmware reboots the partition and loads grub loader
like the normal booting process does. Hence in the FADUMP approach,
the second kernel (after crash) will always use the default initrd
(OS built). So, to support FADUMP, change is required, as in to add
dump capturing steps, in this initrd.
The current kdumpctl script implementation already has the code to
build initrd using mkdumprd. This patch uses the new '--rebuild'
option introduced, in dracut, to incrementally build the initramfs
image. Before rebuilding, we may need to probe the initrd image for
fadump support, to avoid rebuilding the initrd image multiple times
unnecessarily. This can be done using "lsinitrd" tool with the newly
proposed '--mod' option & inspecting the presence of "kdumpbase" in
the list of modules of default initrd image. We rebuild the image if
only "kdumpbase" module is missing in the initrd image. Also, before
rebuilding, a backup of default initrd image is taken.
Kexec-tools package in rhel7 is now enhanced to insert a out-of-tree
kdump module for dracut, which is responsible for adding vmcore
capture steps into initrd, if dracut is invoked with "IN_KDUMP"
environment variable set to 1. mkdumprd script exports "IN_KDUMP=1"
environment variable before invoking dracut to build kdump initrd.
This patch relies on this current mechanism of kdump init script.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
2014-07-24 18:39:07 +00:00
|
|
|
if [ $? != 0 ]; then
|
|
|
|
echo "mkdumprd: failed to rebuild initrd with fadump support" >&2
|
2014-08-04 10:46:36 +00:00
|
|
|
rm -f /tmp/fadump.initramfs
|
kdump: Rebuild default initrd for firmware assisted dump
The current kdump infrastructure builds a separate initrd which then
gets loaded into memory by kexec-tools for use by kdump kernel. But
firmware assisted dump (FADUMP) does not use kexec-based approach.
After crash, firmware reboots the partition and loads grub loader
like the normal booting process does. Hence in the FADUMP approach,
the second kernel (after crash) will always use the default initrd
(OS built). So, to support FADUMP, change is required, as in to add
dump capturing steps, in this initrd.
The current kdumpctl script implementation already has the code to
build initrd using mkdumprd. This patch uses the new '--rebuild'
option introduced, in dracut, to incrementally build the initramfs
image. Before rebuilding, we may need to probe the initrd image for
fadump support, to avoid rebuilding the initrd image multiple times
unnecessarily. This can be done using "lsinitrd" tool with the newly
proposed '--mod' option & inspecting the presence of "kdumpbase" in
the list of modules of default initrd image. We rebuild the image if
only "kdumpbase" module is missing in the initrd image. Also, before
rebuilding, a backup of default initrd image is taken.
Kexec-tools package in rhel7 is now enhanced to insert a out-of-tree
kdump module for dracut, which is responsible for adding vmcore
capture steps into initrd, if dracut is invoked with "IN_KDUMP"
environment variable set to 1. mkdumprd script exports "IN_KDUMP=1"
environment variable before invoking dracut to build kdump initrd.
This patch relies on this current mechanism of kdump init script.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
2014-07-24 18:39:07 +00:00
|
|
|
return 1
|
|
|
|
fi
|
2014-08-04 10:46:36 +00:00
|
|
|
rm -f /tmp/fadump.initramfs
|
kdump: Rebuild default initrd for firmware assisted dump
The current kdump infrastructure builds a separate initrd which then
gets loaded into memory by kexec-tools for use by kdump kernel. But
firmware assisted dump (FADUMP) does not use kexec-based approach.
After crash, firmware reboots the partition and loads grub loader
like the normal booting process does. Hence in the FADUMP approach,
the second kernel (after crash) will always use the default initrd
(OS built). So, to support FADUMP, change is required, as in to add
dump capturing steps, in this initrd.
The current kdumpctl script implementation already has the code to
build initrd using mkdumprd. This patch uses the new '--rebuild'
option introduced, in dracut, to incrementally build the initramfs
image. Before rebuilding, we may need to probe the initrd image for
fadump support, to avoid rebuilding the initrd image multiple times
unnecessarily. This can be done using "lsinitrd" tool with the newly
proposed '--mod' option & inspecting the presence of "kdumpbase" in
the list of modules of default initrd image. We rebuild the image if
only "kdumpbase" module is missing in the initrd image. Also, before
rebuilding, a backup of default initrd image is taken.
Kexec-tools package in rhel7 is now enhanced to insert a out-of-tree
kdump module for dracut, which is responsible for adding vmcore
capture steps into initrd, if dracut is invoked with "IN_KDUMP"
environment variable set to 1. mkdumprd script exports "IN_KDUMP=1"
environment variable before invoking dracut to build kdump initrd.
This patch relies on this current mechanism of kdump init script.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
2014-07-24 18:39:07 +00:00
|
|
|
|
|
|
|
# updating fadump initrd
|
|
|
|
mv $target_initrd_tmp $TARGET_INITRD
|
|
|
|
sync
|
|
|
|
|
|
|
|
return 0
|
|
|
|
}
|
|
|
|
|
2019-05-16 05:09:21 +00:00
|
|
|
check_earlykdump_is_enabled()
|
|
|
|
{
|
|
|
|
grep -q -w "rd.earlykdump" /proc/cmdline
|
|
|
|
return $?
|
|
|
|
}
|
|
|
|
|
kdump: Rebuild default initrd for firmware assisted dump
The current kdump infrastructure builds a separate initrd which then
gets loaded into memory by kexec-tools for use by kdump kernel. But
firmware assisted dump (FADUMP) does not use kexec-based approach.
After crash, firmware reboots the partition and loads grub loader
like the normal booting process does. Hence in the FADUMP approach,
the second kernel (after crash) will always use the default initrd
(OS built). So, to support FADUMP, change is required, as in to add
dump capturing steps, in this initrd.
The current kdumpctl script implementation already has the code to
build initrd using mkdumprd. This patch uses the new '--rebuild'
option introduced, in dracut, to incrementally build the initramfs
image. Before rebuilding, we may need to probe the initrd image for
fadump support, to avoid rebuilding the initrd image multiple times
unnecessarily. This can be done using "lsinitrd" tool with the newly
proposed '--mod' option & inspecting the presence of "kdumpbase" in
the list of modules of default initrd image. We rebuild the image if
only "kdumpbase" module is missing in the initrd image. Also, before
rebuilding, a backup of default initrd image is taken.
Kexec-tools package in rhel7 is now enhanced to insert a out-of-tree
kdump module for dracut, which is responsible for adding vmcore
capture steps into initrd, if dracut is invoked with "IN_KDUMP"
environment variable set to 1. mkdumprd script exports "IN_KDUMP=1"
environment variable before invoking dracut to build kdump initrd.
This patch relies on this current mechanism of kdump init script.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
2014-07-24 18:39:07 +00:00
|
|
|
rebuild_kdump_initrd()
|
|
|
|
{
|
|
|
|
$MKDUMPRD $TARGET_INITRD $kdump_kver
|
2012-06-14 01:57:27 +00:00
|
|
|
if [ $? != 0 ]; then
|
2013-02-18 09:05:41 +00:00
|
|
|
echo "mkdumprd: failed to make kdump initrd" >&2
|
2012-06-14 01:57:27 +00:00
|
|
|
return 1
|
|
|
|
fi
|
kdump: Rebuild default initrd for firmware assisted dump
The current kdump infrastructure builds a separate initrd which then
gets loaded into memory by kexec-tools for use by kdump kernel. But
firmware assisted dump (FADUMP) does not use kexec-based approach.
After crash, firmware reboots the partition and loads grub loader
like the normal booting process does. Hence in the FADUMP approach,
the second kernel (after crash) will always use the default initrd
(OS built). So, to support FADUMP, change is required, as in to add
dump capturing steps, in this initrd.
The current kdumpctl script implementation already has the code to
build initrd using mkdumprd. This patch uses the new '--rebuild'
option introduced, in dracut, to incrementally build the initramfs
image. Before rebuilding, we may need to probe the initrd image for
fadump support, to avoid rebuilding the initrd image multiple times
unnecessarily. This can be done using "lsinitrd" tool with the newly
proposed '--mod' option & inspecting the presence of "kdumpbase" in
the list of modules of default initrd image. We rebuild the image if
only "kdumpbase" module is missing in the initrd image. Also, before
rebuilding, a backup of default initrd image is taken.
Kexec-tools package in rhel7 is now enhanced to insert a out-of-tree
kdump module for dracut, which is responsible for adding vmcore
capture steps into initrd, if dracut is invoked with "IN_KDUMP"
environment variable set to 1. mkdumprd script exports "IN_KDUMP=1"
environment variable before invoking dracut to build kdump initrd.
This patch relies on this current mechanism of kdump init script.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
2014-07-24 18:39:07 +00:00
|
|
|
|
2019-05-16 05:09:21 +00:00
|
|
|
if check_earlykdump_is_enabled; then
|
|
|
|
echo "Tips: If early kdump is enabled, also require rebuilding the system initramfs to make the changes take effect for early kdump."
|
|
|
|
fi
|
|
|
|
|
kdump: Rebuild default initrd for firmware assisted dump
The current kdump infrastructure builds a separate initrd which then
gets loaded into memory by kexec-tools for use by kdump kernel. But
firmware assisted dump (FADUMP) does not use kexec-based approach.
After crash, firmware reboots the partition and loads grub loader
like the normal booting process does. Hence in the FADUMP approach,
the second kernel (after crash) will always use the default initrd
(OS built). So, to support FADUMP, change is required, as in to add
dump capturing steps, in this initrd.
The current kdumpctl script implementation already has the code to
build initrd using mkdumprd. This patch uses the new '--rebuild'
option introduced, in dracut, to incrementally build the initramfs
image. Before rebuilding, we may need to probe the initrd image for
fadump support, to avoid rebuilding the initrd image multiple times
unnecessarily. This can be done using "lsinitrd" tool with the newly
proposed '--mod' option & inspecting the presence of "kdumpbase" in
the list of modules of default initrd image. We rebuild the image if
only "kdumpbase" module is missing in the initrd image. Also, before
rebuilding, a backup of default initrd image is taken.
Kexec-tools package in rhel7 is now enhanced to insert a out-of-tree
kdump module for dracut, which is responsible for adding vmcore
capture steps into initrd, if dracut is invoked with "IN_KDUMP"
environment variable set to 1. mkdumprd script exports "IN_KDUMP=1"
environment variable before invoking dracut to build kdump initrd.
This patch relies on this current mechanism of kdump init script.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
2014-07-24 18:39:07 +00:00
|
|
|
return 0
|
|
|
|
}
|
|
|
|
|
|
|
|
rebuild_initrd()
|
|
|
|
{
|
2019-03-29 03:29:30 +00:00
|
|
|
if [[ ! -w "$KDUMP_BOOTDIR" ]];then
|
|
|
|
echo "$KDUMP_BOOTDIR does not have write permission. Can not rebuild $TARGET_INITRD"
|
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
|
kdump: Rebuild default initrd for firmware assisted dump
The current kdump infrastructure builds a separate initrd which then
gets loaded into memory by kexec-tools for use by kdump kernel. But
firmware assisted dump (FADUMP) does not use kexec-based approach.
After crash, firmware reboots the partition and loads grub loader
like the normal booting process does. Hence in the FADUMP approach,
the second kernel (after crash) will always use the default initrd
(OS built). So, to support FADUMP, change is required, as in to add
dump capturing steps, in this initrd.
The current kdumpctl script implementation already has the code to
build initrd using mkdumprd. This patch uses the new '--rebuild'
option introduced, in dracut, to incrementally build the initramfs
image. Before rebuilding, we may need to probe the initrd image for
fadump support, to avoid rebuilding the initrd image multiple times
unnecessarily. This can be done using "lsinitrd" tool with the newly
proposed '--mod' option & inspecting the presence of "kdumpbase" in
the list of modules of default initrd image. We rebuild the image if
only "kdumpbase" module is missing in the initrd image. Also, before
rebuilding, a backup of default initrd image is taken.
Kexec-tools package in rhel7 is now enhanced to insert a out-of-tree
kdump module for dracut, which is responsible for adding vmcore
capture steps into initrd, if dracut is invoked with "IN_KDUMP"
environment variable set to 1. mkdumprd script exports "IN_KDUMP=1"
environment variable before invoking dracut to build kdump initrd.
This patch relies on this current mechanism of kdump init script.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
2014-07-24 18:39:07 +00:00
|
|
|
if [ $DEFAULT_DUMP_MODE == "fadump" ]; then
|
|
|
|
rebuild_fadump_initrd
|
|
|
|
else
|
|
|
|
rebuild_kdump_initrd
|
|
|
|
fi
|
|
|
|
|
|
|
|
return $?
|
2012-06-14 01:57:27 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
#$1: the files to be checked with IFS=' '
|
2014-07-15 06:41:54 +00:00
|
|
|
check_exist()
|
2012-06-14 01:57:27 +00:00
|
|
|
{
|
|
|
|
for file in $1; do
|
|
|
|
if [ ! -f "$file" ]; then
|
|
|
|
echo -n "Error: $file not found."; echo
|
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
done
|
|
|
|
}
|
|
|
|
|
|
|
|
#$1: the files to be checked with IFS=' '
|
2014-07-15 06:41:54 +00:00
|
|
|
check_executable()
|
2012-06-14 01:57:27 +00:00
|
|
|
{
|
|
|
|
for file in $1; do
|
|
|
|
if [ ! -x "$file" ]; then
|
|
|
|
echo -n "Error: $file is not executable."; echo
|
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
done
|
|
|
|
}
|
|
|
|
|
2016-11-03 18:46:31 +00:00
|
|
|
backup_default_initrd()
|
kdump: Rebuild default initrd for firmware assisted dump
The current kdump infrastructure builds a separate initrd which then
gets loaded into memory by kexec-tools for use by kdump kernel. But
firmware assisted dump (FADUMP) does not use kexec-based approach.
After crash, firmware reboots the partition and loads grub loader
like the normal booting process does. Hence in the FADUMP approach,
the second kernel (after crash) will always use the default initrd
(OS built). So, to support FADUMP, change is required, as in to add
dump capturing steps, in this initrd.
The current kdumpctl script implementation already has the code to
build initrd using mkdumprd. This patch uses the new '--rebuild'
option introduced, in dracut, to incrementally build the initramfs
image. Before rebuilding, we may need to probe the initrd image for
fadump support, to avoid rebuilding the initrd image multiple times
unnecessarily. This can be done using "lsinitrd" tool with the newly
proposed '--mod' option & inspecting the presence of "kdumpbase" in
the list of modules of default initrd image. We rebuild the image if
only "kdumpbase" module is missing in the initrd image. Also, before
rebuilding, a backup of default initrd image is taken.
Kexec-tools package in rhel7 is now enhanced to insert a out-of-tree
kdump module for dracut, which is responsible for adding vmcore
capture steps into initrd, if dracut is invoked with "IN_KDUMP"
environment variable set to 1. mkdumprd script exports "IN_KDUMP=1"
environment variable before invoking dracut to build kdump initrd.
This patch relies on this current mechanism of kdump init script.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
2014-07-24 18:39:07 +00:00
|
|
|
{
|
2017-05-04 06:56:54 +00:00
|
|
|
if [ ! -f "$DEFAULT_INITRD" ]; then
|
|
|
|
return
|
|
|
|
fi
|
|
|
|
|
2016-11-03 18:46:31 +00:00
|
|
|
if [ ! -e $DEFAULT_INITRD_BAK ]; then
|
|
|
|
echo "Backing up $DEFAULT_INITRD before rebuild."
|
|
|
|
# save checksum to verify before restoring
|
|
|
|
sha1sum $DEFAULT_INITRD > $INITRD_CHECKSUM_LOCATION
|
|
|
|
cp $DEFAULT_INITRD $DEFAULT_INITRD_BAK
|
|
|
|
if [ $? -ne 0 ]; then
|
|
|
|
echo "WARNING: failed to backup $DEFAULT_INITRD."
|
|
|
|
rm -f $DEFAULT_INITRD_BAK
|
|
|
|
fi
|
|
|
|
fi
|
|
|
|
}
|
kdump: Rebuild default initrd for firmware assisted dump
The current kdump infrastructure builds a separate initrd which then
gets loaded into memory by kexec-tools for use by kdump kernel. But
firmware assisted dump (FADUMP) does not use kexec-based approach.
After crash, firmware reboots the partition and loads grub loader
like the normal booting process does. Hence in the FADUMP approach,
the second kernel (after crash) will always use the default initrd
(OS built). So, to support FADUMP, change is required, as in to add
dump capturing steps, in this initrd.
The current kdumpctl script implementation already has the code to
build initrd using mkdumprd. This patch uses the new '--rebuild'
option introduced, in dracut, to incrementally build the initramfs
image. Before rebuilding, we may need to probe the initrd image for
fadump support, to avoid rebuilding the initrd image multiple times
unnecessarily. This can be done using "lsinitrd" tool with the newly
proposed '--mod' option & inspecting the presence of "kdumpbase" in
the list of modules of default initrd image. We rebuild the image if
only "kdumpbase" module is missing in the initrd image. Also, before
rebuilding, a backup of default initrd image is taken.
Kexec-tools package in rhel7 is now enhanced to insert a out-of-tree
kdump module for dracut, which is responsible for adding vmcore
capture steps into initrd, if dracut is invoked with "IN_KDUMP"
environment variable set to 1. mkdumprd script exports "IN_KDUMP=1"
environment variable before invoking dracut to build kdump initrd.
This patch relies on this current mechanism of kdump init script.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
2014-07-24 18:39:07 +00:00
|
|
|
|
2016-11-03 18:46:31 +00:00
|
|
|
restore_default_initrd()
|
|
|
|
{
|
|
|
|
# If a backup initrd exists, we must be switching back from
|
|
|
|
# fadump to kdump. Restore the original default initrd.
|
|
|
|
if [ -f $DEFAULT_INITRD_BAK ] && [ -f $INITRD_CHECKSUM_LOCATION ]; then
|
|
|
|
# verify checksum before restoring
|
|
|
|
backup_checksum=`sha1sum $DEFAULT_INITRD_BAK | awk '{ print $1 }'`
|
|
|
|
default_checksum=`cat $INITRD_CHECKSUM_LOCATION | awk '{ print $1 }'`
|
|
|
|
if [ "$default_checksum" != "$backup_checksum" ]; then
|
|
|
|
echo "WARNING: checksum mismatch! Can't restore original initrd.."
|
|
|
|
else
|
|
|
|
rm -f $INITRD_CHECKSUM_LOCATION
|
|
|
|
mv $DEFAULT_INITRD_BAK $DEFAULT_INITRD
|
|
|
|
if [[ $? -eq 0 ]]; then
|
|
|
|
echo -n "Restoring original initrd as fadump mode "
|
|
|
|
echo "is disabled."
|
|
|
|
sync
|
|
|
|
fi
|
|
|
|
fi
|
kdump: Rebuild default initrd for firmware assisted dump
The current kdump infrastructure builds a separate initrd which then
gets loaded into memory by kexec-tools for use by kdump kernel. But
firmware assisted dump (FADUMP) does not use kexec-based approach.
After crash, firmware reboots the partition and loads grub loader
like the normal booting process does. Hence in the FADUMP approach,
the second kernel (after crash) will always use the default initrd
(OS built). So, to support FADUMP, change is required, as in to add
dump capturing steps, in this initrd.
The current kdumpctl script implementation already has the code to
build initrd using mkdumprd. This patch uses the new '--rebuild'
option introduced, in dracut, to incrementally build the initramfs
image. Before rebuilding, we may need to probe the initrd image for
fadump support, to avoid rebuilding the initrd image multiple times
unnecessarily. This can be done using "lsinitrd" tool with the newly
proposed '--mod' option & inspecting the presence of "kdumpbase" in
the list of modules of default initrd image. We rebuild the image if
only "kdumpbase" module is missing in the initrd image. Also, before
rebuilding, a backup of default initrd image is taken.
Kexec-tools package in rhel7 is now enhanced to insert a out-of-tree
kdump module for dracut, which is responsible for adding vmcore
capture steps into initrd, if dracut is invoked with "IN_KDUMP"
environment variable set to 1. mkdumprd script exports "IN_KDUMP=1"
environment variable before invoking dracut to build kdump initrd.
This patch relies on this current mechanism of kdump init script.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
2014-07-24 18:39:07 +00:00
|
|
|
fi
|
|
|
|
}
|
|
|
|
|
2014-07-15 06:41:54 +00:00
|
|
|
check_config()
|
2013-03-11 09:31:25 +00:00
|
|
|
{
|
|
|
|
local nr
|
|
|
|
|
Support special mount information via "dracut_args"
There are some complaints about nfs kdump that users must mount
nfs beforehand, which may cause some overhead to nfs server.
For example, there're thounsands of diskless clients deployed with
nfs dumping, each time the client is boot up, it will trigger
kdump rebuilding so will mount nfs, thus resulting in thousands
of nfs request concurrently imposed on the same nfs server.
We introduce a new way of specifying mount information via the
already-existent "dracut_args" directive(so avoid adding extra
directives in /etc/kdump.conf), we will skip all the filesystem
mounting and checking stuff for it. So it can be used in the
above-mentioned nfs scenario to avoid severe nfs server overhead.
Specifically, if there is any "--mount" information specified via
"dracut_args" in /etc/kdump.conf, always use it as the final mount
without any validation(mounting or checking like mount options,
fs size, etc), so users are expected to ensure its correctness.
NOTE:
-Only one mount target is allowed using "dracut_args" globally.
-Dracut will create <mountpoint> if it doesn't exist in kdump kernel,
<mountpoint> must be specified as an absolute path.
-Users should do a test first and ensure it works because kdump does
not prepare the mount or check all the validity.
Reviewed-by: Pratyush Anand <panand@redhat.com>
Suggested-by: Dave Young <dyoung@redhat.com>
Acked-by: Dave Young <dyoung@redhat.com>
Signed-off-by: Xunlei Pang <xlpang@redhat.com>
2016-08-26 03:23:35 +00:00
|
|
|
nr=$(awk 'BEGIN{cnt=0} /^raw|^ssh[[:blank:]]|^nfs|^ext[234]|^xfs|^btrfs|^minix|^dracut_args .*\-\-mount/{cnt++} END{print cnt}' $KDUMP_CONFIG_FILE)
|
2013-03-11 09:31:25 +00:00
|
|
|
[ $nr -gt 1 ] && {
|
|
|
|
echo "More than one dump targets specified."
|
|
|
|
return 1
|
|
|
|
}
|
|
|
|
|
2018-08-10 00:35:54 +00:00
|
|
|
# Check if path option is set more than once.
|
|
|
|
nr=$(awk 'BEGIN{cnt=0} /^path /{cnt++} END{print cnt}' $KDUMP_CONFIG_FILE)
|
|
|
|
[ $nr -gt 1 ] && {
|
|
|
|
echo "Mutiple paths specifed in $KDUMP_CONFIG_FILE"
|
|
|
|
return 1
|
|
|
|
}
|
|
|
|
|
Support special mount information via "dracut_args"
There are some complaints about nfs kdump that users must mount
nfs beforehand, which may cause some overhead to nfs server.
For example, there're thounsands of diskless clients deployed with
nfs dumping, each time the client is boot up, it will trigger
kdump rebuilding so will mount nfs, thus resulting in thousands
of nfs request concurrently imposed on the same nfs server.
We introduce a new way of specifying mount information via the
already-existent "dracut_args" directive(so avoid adding extra
directives in /etc/kdump.conf), we will skip all the filesystem
mounting and checking stuff for it. So it can be used in the
above-mentioned nfs scenario to avoid severe nfs server overhead.
Specifically, if there is any "--mount" information specified via
"dracut_args" in /etc/kdump.conf, always use it as the final mount
without any validation(mounting or checking like mount options,
fs size, etc), so users are expected to ensure its correctness.
NOTE:
-Only one mount target is allowed using "dracut_args" globally.
-Dracut will create <mountpoint> if it doesn't exist in kdump kernel,
<mountpoint> must be specified as an absolute path.
-Users should do a test first and ensure it works because kdump does
not prepare the mount or check all the validity.
Reviewed-by: Pratyush Anand <panand@redhat.com>
Suggested-by: Dave Young <dyoung@redhat.com>
Acked-by: Dave Young <dyoung@redhat.com>
Signed-off-by: Xunlei Pang <xlpang@redhat.com>
2016-08-26 03:23:35 +00:00
|
|
|
nr=$(grep "^dracut_args .*\-\-mount" $KDUMP_CONFIG_FILE | grep -o "\-\-mount" | wc -l)
|
|
|
|
[ $nr -gt 1 ] && {
|
|
|
|
echo "Multiple mount targets specified in one \"dracut_args\"."
|
|
|
|
return 1
|
|
|
|
}
|
|
|
|
|
2017-09-22 11:00:05 +00:00
|
|
|
# Check if we have any leading spaces (or tabs) before the
|
|
|
|
# variable name in the kdump conf file
|
|
|
|
if grep -E -q '^[[:blank:]]+[a-z]' $KDUMP_CONFIG_FILE; then
|
|
|
|
echo "No whitespaces are allowed before a kdump option name in $KDUMP_CONFIG_FILE"
|
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
|
2017-05-11 05:44:28 +00:00
|
|
|
while read config_opt config_val; do
|
2013-03-11 09:31:25 +00:00
|
|
|
case "$config_opt" in
|
2017-05-11 05:44:28 +00:00
|
|
|
\#* | "")
|
2013-03-11 09:31:25 +00:00
|
|
|
;;
|
2019-01-17 20:31:24 +00:00
|
|
|
raw|ext2|ext3|ext4|minix|btrfs|xfs|nfs|ssh|sshkey|path|core_collector|kdump_post|kdump_pre|extra_bins|extra_modules|failure_action|default|final_action|force_rebuild|force_no_rebuild|dracut_args|fence_kdump_args|fence_kdump_nodes)
|
2017-05-11 05:44:29 +00:00
|
|
|
# remove inline comments after the end of a directive.
|
2013-03-11 09:31:25 +00:00
|
|
|
[ -z "$config_val" ] && {
|
|
|
|
echo "Invalid kdump config value for option $config_opt."
|
|
|
|
return 1;
|
|
|
|
}
|
2020-01-28 19:34:48 +00:00
|
|
|
if [ -d "/proc/device-tree/ibm,opal/dump" ] && [ "$config_opt" == "raw" ]; then
|
|
|
|
echo "WARNING: Won't capture opalcore when 'raw' dump target is used."
|
|
|
|
fi
|
2013-03-11 09:31:25 +00:00
|
|
|
;;
|
2013-03-22 03:55:24 +00:00
|
|
|
net|options|link_delay|disk_timeout|debug_mem_level|blacklist)
|
2013-03-11 09:31:25 +00:00
|
|
|
echo "Deprecated kdump config option: $config_opt. Refer to kdump.conf manpage for alternatives."
|
|
|
|
return 1
|
|
|
|
;;
|
|
|
|
*)
|
|
|
|
echo "Invalid kdump config option $config_opt"
|
|
|
|
return 1;
|
|
|
|
;;
|
|
|
|
esac
|
2019-05-17 08:48:46 +00:00
|
|
|
done <<< "$(read_strip_comments $KDUMP_CONFIG_FILE)"
|
2014-07-15 06:41:54 +00:00
|
|
|
|
2019-01-17 20:31:23 +00:00
|
|
|
check_failure_action_config || return 1
|
2019-01-17 20:31:24 +00:00
|
|
|
check_final_action_config || return 1
|
2015-06-11 08:22:11 +00:00
|
|
|
|
2014-04-02 08:33:47 +00:00
|
|
|
check_fence_kdump_config || return 1
|
|
|
|
|
2013-03-11 09:31:25 +00:00
|
|
|
return 0
|
|
|
|
}
|
|
|
|
|
2014-04-02 08:33:47 +00:00
|
|
|
# get_pcs_cluster_modified_files <image timestamp>
|
|
|
|
# return list of modified file for fence_kdump modified in Pacemaker cluster
|
|
|
|
get_pcs_cluster_modified_files()
|
2013-12-17 07:00:33 +00:00
|
|
|
{
|
2014-04-02 08:33:47 +00:00
|
|
|
local time_stamp
|
|
|
|
local modified_files
|
2013-12-17 07:00:33 +00:00
|
|
|
|
2014-04-02 08:33:47 +00:00
|
|
|
is_generic_fence_kdump && return 1
|
2014-04-02 08:33:43 +00:00
|
|
|
is_pcs_fence_kdump || return 1
|
2013-12-17 07:00:33 +00:00
|
|
|
|
2014-04-02 08:33:47 +00:00
|
|
|
time_stamp=`pcs cluster cib | xmllint --xpath 'string(/cib/@cib-last-written)' - | \
|
|
|
|
xargs -0 date +%s --date`
|
2013-12-17 07:00:33 +00:00
|
|
|
|
2014-04-02 08:33:47 +00:00
|
|
|
if [ -n $time_stamp -a $time_stamp -gt $image_time ]; then
|
|
|
|
modified_files="cluster-cib"
|
2013-12-17 07:00:33 +00:00
|
|
|
fi
|
|
|
|
|
2014-04-02 08:33:47 +00:00
|
|
|
if [ -f $FENCE_KDUMP_CONFIG_FILE ]; then
|
|
|
|
time_stamp=`stat -c "%Y" $FENCE_KDUMP_CONFIG_FILE`
|
|
|
|
if [ "$time_stamp" -gt "$image_time" ]; then
|
|
|
|
modified_files="$modified_files $FENCE_KDUMP_CONFIG_FILE"
|
|
|
|
fi
|
|
|
|
fi
|
|
|
|
|
|
|
|
echo $modified_files
|
2013-12-17 07:00:33 +00:00
|
|
|
}
|
|
|
|
|
2016-11-03 18:46:31 +00:00
|
|
|
setup_initrd()
|
kdump: Rebuild default initrd for firmware assisted dump
The current kdump infrastructure builds a separate initrd which then
gets loaded into memory by kexec-tools for use by kdump kernel. But
firmware assisted dump (FADUMP) does not use kexec-based approach.
After crash, firmware reboots the partition and loads grub loader
like the normal booting process does. Hence in the FADUMP approach,
the second kernel (after crash) will always use the default initrd
(OS built). So, to support FADUMP, change is required, as in to add
dump capturing steps, in this initrd.
The current kdumpctl script implementation already has the code to
build initrd using mkdumprd. This patch uses the new '--rebuild'
option introduced, in dracut, to incrementally build the initramfs
image. Before rebuilding, we may need to probe the initrd image for
fadump support, to avoid rebuilding the initrd image multiple times
unnecessarily. This can be done using "lsinitrd" tool with the newly
proposed '--mod' option & inspecting the presence of "kdumpbase" in
the list of modules of default initrd image. We rebuild the image if
only "kdumpbase" module is missing in the initrd image. Also, before
rebuilding, a backup of default initrd image is taken.
Kexec-tools package in rhel7 is now enhanced to insert a out-of-tree
kdump module for dracut, which is responsible for adding vmcore
capture steps into initrd, if dracut is invoked with "IN_KDUMP"
environment variable set to 1. mkdumprd script exports "IN_KDUMP=1"
environment variable before invoking dracut to build kdump initrd.
This patch relies on this current mechanism of kdump init script.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
2014-07-24 18:39:07 +00:00
|
|
|
{
|
2018-10-23 14:13:28 +00:00
|
|
|
KDUMP_BOOTDIR=$(check_boot_dir "${KDUMP_BOOTDIR}")
|
|
|
|
|
|
|
|
if [ -z "$KDUMP_KERNELVER" ]; then
|
|
|
|
kdump_kver=`uname -r`
|
|
|
|
else
|
|
|
|
kdump_kver=$KDUMP_KERNELVER
|
|
|
|
fi
|
|
|
|
|
|
|
|
kdump_kernel="${KDUMP_BOOTDIR}/${KDUMP_IMG}-${kdump_kver}${KDUMP_IMG_EXT}"
|
|
|
|
|
2016-11-03 18:46:31 +00:00
|
|
|
DEFAULT_INITRD="${KDUMP_BOOTDIR}/initramfs-`uname -r`.img"
|
|
|
|
DEFAULT_INITRD_BAK="${KDUMP_BOOTDIR}/.initramfs-`uname -r`.img.default"
|
kdump: Rebuild default initrd for firmware assisted dump
The current kdump infrastructure builds a separate initrd which then
gets loaded into memory by kexec-tools for use by kdump kernel. But
firmware assisted dump (FADUMP) does not use kexec-based approach.
After crash, firmware reboots the partition and loads grub loader
like the normal booting process does. Hence in the FADUMP approach,
the second kernel (after crash) will always use the default initrd
(OS built). So, to support FADUMP, change is required, as in to add
dump capturing steps, in this initrd.
The current kdumpctl script implementation already has the code to
build initrd using mkdumprd. This patch uses the new '--rebuild'
option introduced, in dracut, to incrementally build the initramfs
image. Before rebuilding, we may need to probe the initrd image for
fadump support, to avoid rebuilding the initrd image multiple times
unnecessarily. This can be done using "lsinitrd" tool with the newly
proposed '--mod' option & inspecting the presence of "kdumpbase" in
the list of modules of default initrd image. We rebuild the image if
only "kdumpbase" module is missing in the initrd image. Also, before
rebuilding, a backup of default initrd image is taken.
Kexec-tools package in rhel7 is now enhanced to insert a out-of-tree
kdump module for dracut, which is responsible for adding vmcore
capture steps into initrd, if dracut is invoked with "IN_KDUMP"
environment variable set to 1. mkdumprd script exports "IN_KDUMP=1"
environment variable before invoking dracut to build kdump initrd.
This patch relies on this current mechanism of kdump init script.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
2014-07-24 18:39:07 +00:00
|
|
|
if [ $DEFAULT_DUMP_MODE == "fadump" ]; then
|
2016-11-03 18:46:31 +00:00
|
|
|
TARGET_INITRD="$DEFAULT_INITRD"
|
2019-03-29 03:29:30 +00:00
|
|
|
|
|
|
|
# backup initrd for reference before replacing it
|
|
|
|
# with fadump aware initrd
|
|
|
|
backup_default_initrd
|
kdump: Rebuild default initrd for firmware assisted dump
The current kdump infrastructure builds a separate initrd which then
gets loaded into memory by kexec-tools for use by kdump kernel. But
firmware assisted dump (FADUMP) does not use kexec-based approach.
After crash, firmware reboots the partition and loads grub loader
like the normal booting process does. Hence in the FADUMP approach,
the second kernel (after crash) will always use the default initrd
(OS built). So, to support FADUMP, change is required, as in to add
dump capturing steps, in this initrd.
The current kdumpctl script implementation already has the code to
build initrd using mkdumprd. This patch uses the new '--rebuild'
option introduced, in dracut, to incrementally build the initramfs
image. Before rebuilding, we may need to probe the initrd image for
fadump support, to avoid rebuilding the initrd image multiple times
unnecessarily. This can be done using "lsinitrd" tool with the newly
proposed '--mod' option & inspecting the presence of "kdumpbase" in
the list of modules of default initrd image. We rebuild the image if
only "kdumpbase" module is missing in the initrd image. Also, before
rebuilding, a backup of default initrd image is taken.
Kexec-tools package in rhel7 is now enhanced to insert a out-of-tree
kdump module for dracut, which is responsible for adding vmcore
capture steps into initrd, if dracut is invoked with "IN_KDUMP"
environment variable set to 1. mkdumprd script exports "IN_KDUMP=1"
environment variable before invoking dracut to build kdump initrd.
This patch relies on this current mechanism of kdump init script.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
2014-07-24 18:39:07 +00:00
|
|
|
else
|
|
|
|
TARGET_INITRD="${KDUMP_BOOTDIR}/initramfs-${kdump_kver}kdump.img"
|
2019-03-29 03:29:30 +00:00
|
|
|
|
|
|
|
# check if a backup of default initrd exists. If yes,
|
|
|
|
# it signifies a switch from fadump mode. So, restore
|
|
|
|
# the backed up default initrd.
|
|
|
|
restore_default_initrd
|
kdump: Rebuild default initrd for firmware assisted dump
The current kdump infrastructure builds a separate initrd which then
gets loaded into memory by kexec-tools for use by kdump kernel. But
firmware assisted dump (FADUMP) does not use kexec-based approach.
After crash, firmware reboots the partition and loads grub loader
like the normal booting process does. Hence in the FADUMP approach,
the second kernel (after crash) will always use the default initrd
(OS built). So, to support FADUMP, change is required, as in to add
dump capturing steps, in this initrd.
The current kdumpctl script implementation already has the code to
build initrd using mkdumprd. This patch uses the new '--rebuild'
option introduced, in dracut, to incrementally build the initramfs
image. Before rebuilding, we may need to probe the initrd image for
fadump support, to avoid rebuilding the initrd image multiple times
unnecessarily. This can be done using "lsinitrd" tool with the newly
proposed '--mod' option & inspecting the presence of "kdumpbase" in
the list of modules of default initrd image. We rebuild the image if
only "kdumpbase" module is missing in the initrd image. Also, before
rebuilding, a backup of default initrd image is taken.
Kexec-tools package in rhel7 is now enhanced to insert a out-of-tree
kdump module for dracut, which is responsible for adding vmcore
capture steps into initrd, if dracut is invoked with "IN_KDUMP"
environment variable set to 1. mkdumprd script exports "IN_KDUMP=1"
environment variable before invoking dracut to build kdump initrd.
This patch relies on this current mechanism of kdump init script.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
2014-07-24 18:39:07 +00:00
|
|
|
fi
|
|
|
|
}
|
|
|
|
|
2016-05-09 08:17:53 +00:00
|
|
|
check_files_modified()
|
|
|
|
{
|
|
|
|
local modified_files=""
|
|
|
|
|
|
|
|
#also rebuild when Pacemaker cluster conf is changed and fence kdump is enabled.
|
|
|
|
modified_files=$(get_pcs_cluster_modified_files)
|
|
|
|
|
|
|
|
EXTRA_BINS=`grep ^kdump_post $KDUMP_CONFIG_FILE | cut -d\ -f2`
|
|
|
|
CHECK_FILES=`grep ^kdump_pre $KDUMP_CONFIG_FILE | cut -d\ -f2`
|
2020-06-05 02:24:07 +00:00
|
|
|
if [ -d /etc/kdump/post.d ]; then
|
|
|
|
for file in /etc/kdump/post.d/*; do
|
|
|
|
if [ -x "$file" ]; then
|
|
|
|
POST_FILES="$POST_FILES $file"
|
|
|
|
fi
|
|
|
|
done
|
|
|
|
fi
|
|
|
|
if [ -d /etc/kdump/pre.d ]; then
|
|
|
|
for file in /etc/kdump/pre.d/*; do
|
|
|
|
if [ -x "$file" ]; then
|
|
|
|
PRE_FILES="$PRE_FILES $file"
|
|
|
|
fi
|
|
|
|
done
|
|
|
|
fi
|
2018-02-27 11:35:47 +00:00
|
|
|
CORE_COLLECTOR=`grep ^core_collector $KDUMP_CONFIG_FILE | cut -d\ -f2`
|
|
|
|
CORE_COLLECTOR=`type -P $CORE_COLLECTOR`
|
2020-06-05 02:24:07 +00:00
|
|
|
EXTRA_BINS="$EXTRA_BINS $CHECK_FILES $POST_FILES $PRE_FILES"
|
2016-05-09 08:17:53 +00:00
|
|
|
CHECK_FILES=`grep ^extra_bins $KDUMP_CONFIG_FILE | cut -d\ -f2-`
|
|
|
|
EXTRA_BINS="$EXTRA_BINS $CHECK_FILES"
|
2018-02-27 11:35:47 +00:00
|
|
|
files="$KDUMP_CONFIG_FILE $kdump_kernel $EXTRA_BINS $CORE_COLLECTOR"
|
2016-09-14 14:42:17 +00:00
|
|
|
[[ -e /etc/fstab ]] && files="$files /etc/fstab"
|
2016-05-09 08:17:53 +00:00
|
|
|
|
2019-05-13 06:20:25 +00:00
|
|
|
# Check for any updated extra module
|
|
|
|
EXTRA_MODULES="$(grep ^extra_modules $KDUMP_CONFIG_FILE | sed 's/^extra_modules\s*//')"
|
|
|
|
if [ -n "$EXTRA_MODULES" ]; then
|
|
|
|
if [ -e /lib/modules/$kdump_kver/modules.dep ]; then
|
|
|
|
files="$files /lib/modules/$kdump_kver/modules.dep"
|
|
|
|
fi
|
|
|
|
for _module in $EXTRA_MODULES; do
|
|
|
|
_module_file="$(modinfo --set-version "$kdump_kver" --filename "$_module" 2>/dev/null)"
|
|
|
|
if [[ $? -eq 0 ]]; then
|
|
|
|
files="$files $_module_file"
|
|
|
|
for _dep_modules in $(modinfo -F depends $_module | tr ',' ' '); do
|
|
|
|
files="$files $(modinfo --set-version "$kdump_kver" --filename $_dep_modules 2>/dev/null)"
|
|
|
|
done
|
|
|
|
else
|
|
|
|
# If it's not a module nor builtin, give an error
|
|
|
|
if ! ( modprobe --set-version "$kdump_kver" --dry-run "$_module" &>/dev/null ); then
|
|
|
|
echo "Module $_module not found"
|
|
|
|
fi
|
|
|
|
fi
|
|
|
|
done
|
|
|
|
fi
|
|
|
|
|
2016-05-09 08:17:53 +00:00
|
|
|
check_exist "$files" && check_executable "$EXTRA_BINS"
|
|
|
|
[ $? -ne 0 ] && return 2
|
|
|
|
|
|
|
|
for file in $files; do
|
2019-05-13 06:20:24 +00:00
|
|
|
if [ -e "$file" ]; then
|
|
|
|
time_stamp=`stat -c "%Y" $file`
|
|
|
|
if [ "$time_stamp" -gt "$image_time" ]; then
|
|
|
|
modified_files="$modified_files $file"
|
|
|
|
fi
|
|
|
|
if [ -L "$file" ]; then
|
|
|
|
file=$(readlink -m $file)
|
|
|
|
time_stamp=`stat -c "%Y" $file`
|
|
|
|
if [ "$time_stamp" -gt "$image_time" ]; then
|
|
|
|
modified_files="$modified_files $file"
|
|
|
|
fi
|
|
|
|
fi
|
|
|
|
else
|
|
|
|
echo "$file doesn't exist"
|
2016-05-09 08:17:53 +00:00
|
|
|
fi
|
|
|
|
done
|
2019-05-13 06:20:24 +00:00
|
|
|
|
2016-05-09 08:17:53 +00:00
|
|
|
if [ -n "$modified_files" ]; then
|
|
|
|
echo "Detected change(s) in the following file(s):"
|
|
|
|
echo -n " "; echo "$modified_files" | sed 's/\s/\n /g'
|
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
|
|
|
|
return 0
|
|
|
|
}
|
|
|
|
|
kdumpctl: force rebuild in case of file system is modified
kdumpctl passes --device argument if dump target is a raw device. It passes
--mount argument if dump target is either mounted as nfs or as a bulk
device. When dump target device is a root device then it does not pass any
of the above two arguments.
After kdumpctl restart, if there is any change in file system which needs
different dracut arguments, then initramfs must be rebuild.
Modification in filesystem for a raw target does not affect dracut
arguments. So, we do not consider to check any modification if raw target
was specified in kdump.conf.
We might need to change dracut arguments if there is some changes in nfs
and ssh target related files. However, we do not consider them in this
patch.
We mainly consider changes in bulk target specified in kdump.conf. We also
consider changes in bulk and nfs file system, if there was no dump target
specified in kdump.conf but dump path is mounting such file systems.
So the initramfs must be rebuild if, either dump target's persistent path
or it's mount point or its file system type changes. If there is no dump
target specified then, both dump path and root path must mount same device,
otherwise rebuild should be triggered.
Some of the examples when we can need a rebuild:
-- "dump target" is specified as one of ext[234], xfs or btrfs. But after
kdump initramfs building its UUID is changed by reformatting.
-- "dump target" is specified as file system type fs1 (say ext3). But after
kdump initramfs building, user change it to fs2 (say ext4), probably by
a mkfs.ext4 executing on the target device.
-- "dump target" is not specified, but "dump path" mounts a device which
is different than device for root path and either UUID or file system type
is modified after kdump initramfs build.
-- "dump target" is not specified, but "dump path" mounts a nfs device and
nfs host path changes after kdump initramfs build.
Some testing:
Initial conditions:
-- No dump target specified
-- dump path (/var/crash) and root(/) are on same device
-- kdumpctl was already executed once after last modification in
/etc/kdump.conf
# kdumpctl restart
No rebuild
# mkfs.ext2 /dev/md0;mount /dev/md0 /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# mount /dev/md0 /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.ext4 /dev/md0;
# mount /dev/md0 /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.ext4 /dev/mapper/fedora-swap
# mount /dev/mapper/fedora-swap /var/crash/
# kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.btrfs /dev/mapper/fedora-swap -f
# mount /dev/mapper/fedora-swap /var/crash/
# kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# mount /dev/mapper/fedora-swap /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.minix /dev/md0
# mount /dev/md0 /var/crash/; kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# mount 192.168.1.16:/nfsroot /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# mount 192.168.1.16:/nfsroot /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash;mount 192.168.1.12:/nfsroot /var/crash/
# kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
Added "raw /dev/md0" in /etc/kdump.conf
# kdumpctl restart
Rebuilt because "Detected change in /etc/kdump.conf"
# mkfs.ext4 /dev/md0 ;kdumpctl restart
No rebuild
Added "ext4 /dev/md0" in /etc/kdump.conf
# mkfs.ext4 /dev/md0;mount /dev/md0 /mnt
# mkdir /mnt/var;mkdir /mnt/var/crash; kdumpctl restart
Rebuilt because "Detected change in /etc/kdump.conf"
# umount /mnt;mkfs.ext4 /dev/md0;mount /dev/md0 /mnt
# mkdir /mnt/var;mkdir /mnt/var/crash; kdumpctl restart
Rebuilt because "Detected change in /etc/kdump.conf"
Most of the credits for this patch goes to Xunlei Pang <xpang@redhat.com>
for suggesting several improvements.
Signed-off-by: Pratyush Anand <panand@redhat.com>
Acked-by: Xunlei Pang <xlpang@redhat.com>
Acked-by: Baoquan He <xlpang@redhat.com>
Acked-by: Dave Young <dyoung@redhat.com>
2016-05-09 08:17:55 +00:00
|
|
|
check_dump_fs_modified()
|
|
|
|
{
|
|
|
|
local _old_dev _old_mntpoint _old_fstype
|
|
|
|
local _new_dev _new_mntpoint _new_fstype
|
|
|
|
local _target _path _dracut_args
|
2020-03-18 07:51:21 +00:00
|
|
|
local _target_drivers _module_name _module_filename
|
kdumpctl: Detect block device driver change for initramfs rebuild
Previous we rebuild the initramfs when kenrel load module list changed,
but this is not very stable as some async services may load/unload
kernel modules, and cause unnecessary initramfs rebuild.
Instead, it's better to just check if the module required to dump to
the dump target is loaded or not, and rebuild if not loaded. This
avoids most false-positives, and ensure local target change is always
covered.
Currently only local fs dump target is covered, because this check
requires the dump target to be mounted when building the initramfs,
this guarantee that the module is in the loaded kernel module list,
else we may still get some false positive.
dracut-install could be leveraged to combine the modalias list with
kernel loaded module list as a more stable module list in the initramfs,
but upstream dracut change need to be done first.
Passed test on a KVM VM, changing the storage between SATA/USB/VirtIO
will trigger initramfs rebuild and didn't notice any false-positive.
Also passed test on my laptop with no false-positive.
Signed-off-by: Kairui Song <kasong@redhat.com>
Acked-by: Dave Young <dyoung@redhat.com>
2019-05-07 06:19:18 +00:00
|
|
|
|
2020-03-16 15:18:38 +00:00
|
|
|
local _old_drivers="$(lsinitrd $TARGET_INITRD -f /usr/lib/dracut/hostonly-kernel-modules.txt | tr '\n' ' ')"
|
kdumpctl: force rebuild in case of file system is modified
kdumpctl passes --device argument if dump target is a raw device. It passes
--mount argument if dump target is either mounted as nfs or as a bulk
device. When dump target device is a root device then it does not pass any
of the above two arguments.
After kdumpctl restart, if there is any change in file system which needs
different dracut arguments, then initramfs must be rebuild.
Modification in filesystem for a raw target does not affect dracut
arguments. So, we do not consider to check any modification if raw target
was specified in kdump.conf.
We might need to change dracut arguments if there is some changes in nfs
and ssh target related files. However, we do not consider them in this
patch.
We mainly consider changes in bulk target specified in kdump.conf. We also
consider changes in bulk and nfs file system, if there was no dump target
specified in kdump.conf but dump path is mounting such file systems.
So the initramfs must be rebuild if, either dump target's persistent path
or it's mount point or its file system type changes. If there is no dump
target specified then, both dump path and root path must mount same device,
otherwise rebuild should be triggered.
Some of the examples when we can need a rebuild:
-- "dump target" is specified as one of ext[234], xfs or btrfs. But after
kdump initramfs building its UUID is changed by reformatting.
-- "dump target" is specified as file system type fs1 (say ext3). But after
kdump initramfs building, user change it to fs2 (say ext4), probably by
a mkfs.ext4 executing on the target device.
-- "dump target" is not specified, but "dump path" mounts a device which
is different than device for root path and either UUID or file system type
is modified after kdump initramfs build.
-- "dump target" is not specified, but "dump path" mounts a nfs device and
nfs host path changes after kdump initramfs build.
Some testing:
Initial conditions:
-- No dump target specified
-- dump path (/var/crash) and root(/) are on same device
-- kdumpctl was already executed once after last modification in
/etc/kdump.conf
# kdumpctl restart
No rebuild
# mkfs.ext2 /dev/md0;mount /dev/md0 /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# mount /dev/md0 /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.ext4 /dev/md0;
# mount /dev/md0 /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.ext4 /dev/mapper/fedora-swap
# mount /dev/mapper/fedora-swap /var/crash/
# kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.btrfs /dev/mapper/fedora-swap -f
# mount /dev/mapper/fedora-swap /var/crash/
# kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# mount /dev/mapper/fedora-swap /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.minix /dev/md0
# mount /dev/md0 /var/crash/; kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# mount 192.168.1.16:/nfsroot /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# mount 192.168.1.16:/nfsroot /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash;mount 192.168.1.12:/nfsroot /var/crash/
# kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
Added "raw /dev/md0" in /etc/kdump.conf
# kdumpctl restart
Rebuilt because "Detected change in /etc/kdump.conf"
# mkfs.ext4 /dev/md0 ;kdumpctl restart
No rebuild
Added "ext4 /dev/md0" in /etc/kdump.conf
# mkfs.ext4 /dev/md0;mount /dev/md0 /mnt
# mkdir /mnt/var;mkdir /mnt/var/crash; kdumpctl restart
Rebuilt because "Detected change in /etc/kdump.conf"
# umount /mnt;mkfs.ext4 /dev/md0;mount /dev/md0 /mnt
# mkdir /mnt/var;mkdir /mnt/var/crash; kdumpctl restart
Rebuilt because "Detected change in /etc/kdump.conf"
Most of the credits for this patch goes to Xunlei Pang <xpang@redhat.com>
for suggesting several improvements.
Signed-off-by: Pratyush Anand <panand@redhat.com>
Acked-by: Xunlei Pang <xlpang@redhat.com>
Acked-by: Baoquan He <xlpang@redhat.com>
Acked-by: Dave Young <dyoung@redhat.com>
2016-05-09 08:17:55 +00:00
|
|
|
|
Support special mount information via "dracut_args"
There are some complaints about nfs kdump that users must mount
nfs beforehand, which may cause some overhead to nfs server.
For example, there're thounsands of diskless clients deployed with
nfs dumping, each time the client is boot up, it will trigger
kdump rebuilding so will mount nfs, thus resulting in thousands
of nfs request concurrently imposed on the same nfs server.
We introduce a new way of specifying mount information via the
already-existent "dracut_args" directive(so avoid adding extra
directives in /etc/kdump.conf), we will skip all the filesystem
mounting and checking stuff for it. So it can be used in the
above-mentioned nfs scenario to avoid severe nfs server overhead.
Specifically, if there is any "--mount" information specified via
"dracut_args" in /etc/kdump.conf, always use it as the final mount
without any validation(mounting or checking like mount options,
fs size, etc), so users are expected to ensure its correctness.
NOTE:
-Only one mount target is allowed using "dracut_args" globally.
-Dracut will create <mountpoint> if it doesn't exist in kdump kernel,
<mountpoint> must be specified as an absolute path.
-Users should do a test first and ensure it works because kdump does
not prepare the mount or check all the validity.
Reviewed-by: Pratyush Anand <panand@redhat.com>
Suggested-by: Dave Young <dyoung@redhat.com>
Acked-by: Dave Young <dyoung@redhat.com>
Signed-off-by: Xunlei Pang <xlpang@redhat.com>
2016-08-26 03:23:35 +00:00
|
|
|
# No need to check in case of mount target specified via "dracut_args".
|
|
|
|
if is_mount_in_dracut_args; then
|
|
|
|
return 0
|
|
|
|
fi
|
|
|
|
|
kdumpctl: force rebuild in case of file system is modified
kdumpctl passes --device argument if dump target is a raw device. It passes
--mount argument if dump target is either mounted as nfs or as a bulk
device. When dump target device is a root device then it does not pass any
of the above two arguments.
After kdumpctl restart, if there is any change in file system which needs
different dracut arguments, then initramfs must be rebuild.
Modification in filesystem for a raw target does not affect dracut
arguments. So, we do not consider to check any modification if raw target
was specified in kdump.conf.
We might need to change dracut arguments if there is some changes in nfs
and ssh target related files. However, we do not consider them in this
patch.
We mainly consider changes in bulk target specified in kdump.conf. We also
consider changes in bulk and nfs file system, if there was no dump target
specified in kdump.conf but dump path is mounting such file systems.
So the initramfs must be rebuild if, either dump target's persistent path
or it's mount point or its file system type changes. If there is no dump
target specified then, both dump path and root path must mount same device,
otherwise rebuild should be triggered.
Some of the examples when we can need a rebuild:
-- "dump target" is specified as one of ext[234], xfs or btrfs. But after
kdump initramfs building its UUID is changed by reformatting.
-- "dump target" is specified as file system type fs1 (say ext3). But after
kdump initramfs building, user change it to fs2 (say ext4), probably by
a mkfs.ext4 executing on the target device.
-- "dump target" is not specified, but "dump path" mounts a device which
is different than device for root path and either UUID or file system type
is modified after kdump initramfs build.
-- "dump target" is not specified, but "dump path" mounts a nfs device and
nfs host path changes after kdump initramfs build.
Some testing:
Initial conditions:
-- No dump target specified
-- dump path (/var/crash) and root(/) are on same device
-- kdumpctl was already executed once after last modification in
/etc/kdump.conf
# kdumpctl restart
No rebuild
# mkfs.ext2 /dev/md0;mount /dev/md0 /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# mount /dev/md0 /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.ext4 /dev/md0;
# mount /dev/md0 /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.ext4 /dev/mapper/fedora-swap
# mount /dev/mapper/fedora-swap /var/crash/
# kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.btrfs /dev/mapper/fedora-swap -f
# mount /dev/mapper/fedora-swap /var/crash/
# kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# mount /dev/mapper/fedora-swap /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.minix /dev/md0
# mount /dev/md0 /var/crash/; kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# mount 192.168.1.16:/nfsroot /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# mount 192.168.1.16:/nfsroot /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash;mount 192.168.1.12:/nfsroot /var/crash/
# kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
Added "raw /dev/md0" in /etc/kdump.conf
# kdumpctl restart
Rebuilt because "Detected change in /etc/kdump.conf"
# mkfs.ext4 /dev/md0 ;kdumpctl restart
No rebuild
Added "ext4 /dev/md0" in /etc/kdump.conf
# mkfs.ext4 /dev/md0;mount /dev/md0 /mnt
# mkdir /mnt/var;mkdir /mnt/var/crash; kdumpctl restart
Rebuilt because "Detected change in /etc/kdump.conf"
# umount /mnt;mkfs.ext4 /dev/md0;mount /dev/md0 /mnt
# mkdir /mnt/var;mkdir /mnt/var/crash; kdumpctl restart
Rebuilt because "Detected change in /etc/kdump.conf"
Most of the credits for this patch goes to Xunlei Pang <xpang@redhat.com>
for suggesting several improvements.
Signed-off-by: Pratyush Anand <panand@redhat.com>
Acked-by: Xunlei Pang <xlpang@redhat.com>
Acked-by: Baoquan He <xlpang@redhat.com>
Acked-by: Dave Young <dyoung@redhat.com>
2016-05-09 08:17:55 +00:00
|
|
|
# No need to check in case of raw target.
|
|
|
|
# Currently we do not check also if ssh/nfs target is specified
|
|
|
|
if is_ssh_dump_target || is_nfs_dump_target || is_raw_dump_target; then
|
|
|
|
return 0
|
|
|
|
fi
|
|
|
|
|
|
|
|
_target=$(get_user_configured_dump_disk)
|
|
|
|
|
|
|
|
if [[ -n "$_target" ]]; then
|
|
|
|
_target=$(to_dev_name $_target)
|
|
|
|
_new_fstype=$(blkid $_target | awk -F"TYPE=" '{print $2}' | cut -d '"' -f 2)
|
|
|
|
else
|
|
|
|
_path=$(get_save_path)
|
2016-09-12 07:10:04 +00:00
|
|
|
_target=$(get_target_from_path $_path)
|
|
|
|
_target=$(to_dev_name $_target)
|
|
|
|
_new_fstype=$(get_fs_type_from_target $_target)
|
kdumpctl: force rebuild in case of file system is modified
kdumpctl passes --device argument if dump target is a raw device. It passes
--mount argument if dump target is either mounted as nfs or as a bulk
device. When dump target device is a root device then it does not pass any
of the above two arguments.
After kdumpctl restart, if there is any change in file system which needs
different dracut arguments, then initramfs must be rebuild.
Modification in filesystem for a raw target does not affect dracut
arguments. So, we do not consider to check any modification if raw target
was specified in kdump.conf.
We might need to change dracut arguments if there is some changes in nfs
and ssh target related files. However, we do not consider them in this
patch.
We mainly consider changes in bulk target specified in kdump.conf. We also
consider changes in bulk and nfs file system, if there was no dump target
specified in kdump.conf but dump path is mounting such file systems.
So the initramfs must be rebuild if, either dump target's persistent path
or it's mount point or its file system type changes. If there is no dump
target specified then, both dump path and root path must mount same device,
otherwise rebuild should be triggered.
Some of the examples when we can need a rebuild:
-- "dump target" is specified as one of ext[234], xfs or btrfs. But after
kdump initramfs building its UUID is changed by reformatting.
-- "dump target" is specified as file system type fs1 (say ext3). But after
kdump initramfs building, user change it to fs2 (say ext4), probably by
a mkfs.ext4 executing on the target device.
-- "dump target" is not specified, but "dump path" mounts a device which
is different than device for root path and either UUID or file system type
is modified after kdump initramfs build.
-- "dump target" is not specified, but "dump path" mounts a nfs device and
nfs host path changes after kdump initramfs build.
Some testing:
Initial conditions:
-- No dump target specified
-- dump path (/var/crash) and root(/) are on same device
-- kdumpctl was already executed once after last modification in
/etc/kdump.conf
# kdumpctl restart
No rebuild
# mkfs.ext2 /dev/md0;mount /dev/md0 /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# mount /dev/md0 /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.ext4 /dev/md0;
# mount /dev/md0 /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.ext4 /dev/mapper/fedora-swap
# mount /dev/mapper/fedora-swap /var/crash/
# kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.btrfs /dev/mapper/fedora-swap -f
# mount /dev/mapper/fedora-swap /var/crash/
# kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# mount /dev/mapper/fedora-swap /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.minix /dev/md0
# mount /dev/md0 /var/crash/; kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# mount 192.168.1.16:/nfsroot /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# mount 192.168.1.16:/nfsroot /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash;mount 192.168.1.12:/nfsroot /var/crash/
# kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
Added "raw /dev/md0" in /etc/kdump.conf
# kdumpctl restart
Rebuilt because "Detected change in /etc/kdump.conf"
# mkfs.ext4 /dev/md0 ;kdumpctl restart
No rebuild
Added "ext4 /dev/md0" in /etc/kdump.conf
# mkfs.ext4 /dev/md0;mount /dev/md0 /mnt
# mkdir /mnt/var;mkdir /mnt/var/crash; kdumpctl restart
Rebuilt because "Detected change in /etc/kdump.conf"
# umount /mnt;mkfs.ext4 /dev/md0;mount /dev/md0 /mnt
# mkdir /mnt/var;mkdir /mnt/var/crash; kdumpctl restart
Rebuilt because "Detected change in /etc/kdump.conf"
Most of the credits for this patch goes to Xunlei Pang <xpang@redhat.com>
for suggesting several improvements.
Signed-off-by: Pratyush Anand <panand@redhat.com>
Acked-by: Xunlei Pang <xlpang@redhat.com>
Acked-by: Baoquan He <xlpang@redhat.com>
Acked-by: Dave Young <dyoung@redhat.com>
2016-05-09 08:17:55 +00:00
|
|
|
if [[ -z "$_target" || -z "$_new_fstype" ]];then
|
|
|
|
echo "Dump path $_path does not exist"
|
|
|
|
return 2
|
|
|
|
fi
|
|
|
|
fi
|
|
|
|
|
kdumpctl: Detect block device driver change for initramfs rebuild
Previous we rebuild the initramfs when kenrel load module list changed,
but this is not very stable as some async services may load/unload
kernel modules, and cause unnecessary initramfs rebuild.
Instead, it's better to just check if the module required to dump to
the dump target is loaded or not, and rebuild if not loaded. This
avoids most false-positives, and ensure local target change is always
covered.
Currently only local fs dump target is covered, because this check
requires the dump target to be mounted when building the initramfs,
this guarantee that the module is in the loaded kernel module list,
else we may still get some false positive.
dracut-install could be leveraged to combine the modalias list with
kernel loaded module list as a more stable module list in the initramfs,
but upstream dracut change need to be done first.
Passed test on a KVM VM, changing the storage between SATA/USB/VirtIO
will trigger initramfs rebuild and didn't notice any false-positive.
Also passed test on my laptop with no false-positive.
Signed-off-by: Kairui Song <kasong@redhat.com>
Acked-by: Dave Young <dyoung@redhat.com>
2019-05-07 06:19:18 +00:00
|
|
|
_record_block_drivers() {
|
|
|
|
local _drivers
|
|
|
|
if [[ -b /dev/block/$1 ]]; then
|
|
|
|
_drivers=$(udevadm info -a "/dev/block/$1" | sed -n 's/\s*DRIVERS=="\(\S\+\)"/\1/p')
|
|
|
|
fi
|
|
|
|
if [[ -b $1 ]]; then
|
|
|
|
_drivers=$(udevadm info -a "$1" | sed -n 's/\s*DRIVERS=="\(\S\+\)"/\1/p')
|
|
|
|
fi
|
|
|
|
for _driver in $_drivers; do
|
|
|
|
if ! [[ " $_target_drivers " == *" $_driver "* ]]; then
|
|
|
|
_target_drivers="$_target_drivers $_driver"
|
|
|
|
fi
|
|
|
|
done
|
|
|
|
return 1
|
|
|
|
}
|
|
|
|
|
|
|
|
check_block_and_slaves_all _record_block_drivers "$(get_maj_min "$_target")"
|
|
|
|
for _driver in $_target_drivers; do
|
2020-03-18 07:51:21 +00:00
|
|
|
# Skip deprecated/invalid driver name or built-in module
|
kdumpctl: Detect block device driver change for initramfs rebuild
Previous we rebuild the initramfs when kenrel load module list changed,
but this is not very stable as some async services may load/unload
kernel modules, and cause unnecessary initramfs rebuild.
Instead, it's better to just check if the module required to dump to
the dump target is loaded or not, and rebuild if not loaded. This
avoids most false-positives, and ensure local target change is always
covered.
Currently only local fs dump target is covered, because this check
requires the dump target to be mounted when building the initramfs,
this guarantee that the module is in the loaded kernel module list,
else we may still get some false positive.
dracut-install could be leveraged to combine the modalias list with
kernel loaded module list as a more stable module list in the initramfs,
but upstream dracut change need to be done first.
Passed test on a KVM VM, changing the storage between SATA/USB/VirtIO
will trigger initramfs rebuild and didn't notice any false-positive.
Also passed test on my laptop with no false-positive.
Signed-off-by: Kairui Song <kasong@redhat.com>
Acked-by: Dave Young <dyoung@redhat.com>
2019-05-07 06:19:18 +00:00
|
|
|
_module_name=$(modinfo --set-version "$kdump_kver" -F name $_driver 2>/dev/null)
|
2020-03-18 07:51:21 +00:00
|
|
|
_module_filename=$(modinfo --set-version "$kdump_kver" -n $_driver 2>/dev/null)
|
|
|
|
if [ $? -ne 0 ] || [ -z "$_module_name" ] || [[ "$_module_filename" = *"(builtin)"* ]]; then
|
kdumpctl: Detect block device driver change for initramfs rebuild
Previous we rebuild the initramfs when kenrel load module list changed,
but this is not very stable as some async services may load/unload
kernel modules, and cause unnecessary initramfs rebuild.
Instead, it's better to just check if the module required to dump to
the dump target is loaded or not, and rebuild if not loaded. This
avoids most false-positives, and ensure local target change is always
covered.
Currently only local fs dump target is covered, because this check
requires the dump target to be mounted when building the initramfs,
this guarantee that the module is in the loaded kernel module list,
else we may still get some false positive.
dracut-install could be leveraged to combine the modalias list with
kernel loaded module list as a more stable module list in the initramfs,
but upstream dracut change need to be done first.
Passed test on a KVM VM, changing the storage between SATA/USB/VirtIO
will trigger initramfs rebuild and didn't notice any false-positive.
Also passed test on my laptop with no false-positive.
Signed-off-by: Kairui Song <kasong@redhat.com>
Acked-by: Dave Young <dyoung@redhat.com>
2019-05-07 06:19:18 +00:00
|
|
|
continue
|
|
|
|
fi
|
|
|
|
if ! [[ " $_old_drivers " == *" $_module_name "* ]]; then
|
|
|
|
echo "Detected change in block device driver, new loaded module: $_module_name"
|
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
done
|
|
|
|
|
kdumpctl: force rebuild in case of file system is modified
kdumpctl passes --device argument if dump target is a raw device. It passes
--mount argument if dump target is either mounted as nfs or as a bulk
device. When dump target device is a root device then it does not pass any
of the above two arguments.
After kdumpctl restart, if there is any change in file system which needs
different dracut arguments, then initramfs must be rebuild.
Modification in filesystem for a raw target does not affect dracut
arguments. So, we do not consider to check any modification if raw target
was specified in kdump.conf.
We might need to change dracut arguments if there is some changes in nfs
and ssh target related files. However, we do not consider them in this
patch.
We mainly consider changes in bulk target specified in kdump.conf. We also
consider changes in bulk and nfs file system, if there was no dump target
specified in kdump.conf but dump path is mounting such file systems.
So the initramfs must be rebuild if, either dump target's persistent path
or it's mount point or its file system type changes. If there is no dump
target specified then, both dump path and root path must mount same device,
otherwise rebuild should be triggered.
Some of the examples when we can need a rebuild:
-- "dump target" is specified as one of ext[234], xfs or btrfs. But after
kdump initramfs building its UUID is changed by reformatting.
-- "dump target" is specified as file system type fs1 (say ext3). But after
kdump initramfs building, user change it to fs2 (say ext4), probably by
a mkfs.ext4 executing on the target device.
-- "dump target" is not specified, but "dump path" mounts a device which
is different than device for root path and either UUID or file system type
is modified after kdump initramfs build.
-- "dump target" is not specified, but "dump path" mounts a nfs device and
nfs host path changes after kdump initramfs build.
Some testing:
Initial conditions:
-- No dump target specified
-- dump path (/var/crash) and root(/) are on same device
-- kdumpctl was already executed once after last modification in
/etc/kdump.conf
# kdumpctl restart
No rebuild
# mkfs.ext2 /dev/md0;mount /dev/md0 /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# mount /dev/md0 /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.ext4 /dev/md0;
# mount /dev/md0 /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.ext4 /dev/mapper/fedora-swap
# mount /dev/mapper/fedora-swap /var/crash/
# kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.btrfs /dev/mapper/fedora-swap -f
# mount /dev/mapper/fedora-swap /var/crash/
# kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# mount /dev/mapper/fedora-swap /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.minix /dev/md0
# mount /dev/md0 /var/crash/; kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# mount 192.168.1.16:/nfsroot /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# mount 192.168.1.16:/nfsroot /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash;mount 192.168.1.12:/nfsroot /var/crash/
# kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
Added "raw /dev/md0" in /etc/kdump.conf
# kdumpctl restart
Rebuilt because "Detected change in /etc/kdump.conf"
# mkfs.ext4 /dev/md0 ;kdumpctl restart
No rebuild
Added "ext4 /dev/md0" in /etc/kdump.conf
# mkfs.ext4 /dev/md0;mount /dev/md0 /mnt
# mkdir /mnt/var;mkdir /mnt/var/crash; kdumpctl restart
Rebuilt because "Detected change in /etc/kdump.conf"
# umount /mnt;mkfs.ext4 /dev/md0;mount /dev/md0 /mnt
# mkdir /mnt/var;mkdir /mnt/var/crash; kdumpctl restart
Rebuilt because "Detected change in /etc/kdump.conf"
Most of the credits for this patch goes to Xunlei Pang <xpang@redhat.com>
for suggesting several improvements.
Signed-off-by: Pratyush Anand <panand@redhat.com>
Acked-by: Xunlei Pang <xlpang@redhat.com>
Acked-by: Baoquan He <xlpang@redhat.com>
Acked-by: Dave Young <dyoung@redhat.com>
2016-05-09 08:17:55 +00:00
|
|
|
if [[ $(expr substr $_new_fstype 1 3) = "nfs" ]];then
|
|
|
|
_new_dev=$_target
|
|
|
|
else
|
2016-11-17 04:51:12 +00:00
|
|
|
_new_dev=$(get_persistent_dev $_target)
|
|
|
|
if [ -z "$_new_dev" ]; then
|
|
|
|
echo "Get persistent device name failed"
|
|
|
|
return 2
|
|
|
|
fi
|
kdumpctl: force rebuild in case of file system is modified
kdumpctl passes --device argument if dump target is a raw device. It passes
--mount argument if dump target is either mounted as nfs or as a bulk
device. When dump target device is a root device then it does not pass any
of the above two arguments.
After kdumpctl restart, if there is any change in file system which needs
different dracut arguments, then initramfs must be rebuild.
Modification in filesystem for a raw target does not affect dracut
arguments. So, we do not consider to check any modification if raw target
was specified in kdump.conf.
We might need to change dracut arguments if there is some changes in nfs
and ssh target related files. However, we do not consider them in this
patch.
We mainly consider changes in bulk target specified in kdump.conf. We also
consider changes in bulk and nfs file system, if there was no dump target
specified in kdump.conf but dump path is mounting such file systems.
So the initramfs must be rebuild if, either dump target's persistent path
or it's mount point or its file system type changes. If there is no dump
target specified then, both dump path and root path must mount same device,
otherwise rebuild should be triggered.
Some of the examples when we can need a rebuild:
-- "dump target" is specified as one of ext[234], xfs or btrfs. But after
kdump initramfs building its UUID is changed by reformatting.
-- "dump target" is specified as file system type fs1 (say ext3). But after
kdump initramfs building, user change it to fs2 (say ext4), probably by
a mkfs.ext4 executing on the target device.
-- "dump target" is not specified, but "dump path" mounts a device which
is different than device for root path and either UUID or file system type
is modified after kdump initramfs build.
-- "dump target" is not specified, but "dump path" mounts a nfs device and
nfs host path changes after kdump initramfs build.
Some testing:
Initial conditions:
-- No dump target specified
-- dump path (/var/crash) and root(/) are on same device
-- kdumpctl was already executed once after last modification in
/etc/kdump.conf
# kdumpctl restart
No rebuild
# mkfs.ext2 /dev/md0;mount /dev/md0 /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# mount /dev/md0 /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.ext4 /dev/md0;
# mount /dev/md0 /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.ext4 /dev/mapper/fedora-swap
# mount /dev/mapper/fedora-swap /var/crash/
# kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.btrfs /dev/mapper/fedora-swap -f
# mount /dev/mapper/fedora-swap /var/crash/
# kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# mount /dev/mapper/fedora-swap /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.minix /dev/md0
# mount /dev/md0 /var/crash/; kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# mount 192.168.1.16:/nfsroot /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# mount 192.168.1.16:/nfsroot /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash;mount 192.168.1.12:/nfsroot /var/crash/
# kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
Added "raw /dev/md0" in /etc/kdump.conf
# kdumpctl restart
Rebuilt because "Detected change in /etc/kdump.conf"
# mkfs.ext4 /dev/md0 ;kdumpctl restart
No rebuild
Added "ext4 /dev/md0" in /etc/kdump.conf
# mkfs.ext4 /dev/md0;mount /dev/md0 /mnt
# mkdir /mnt/var;mkdir /mnt/var/crash; kdumpctl restart
Rebuilt because "Detected change in /etc/kdump.conf"
# umount /mnt;mkfs.ext4 /dev/md0;mount /dev/md0 /mnt
# mkdir /mnt/var;mkdir /mnt/var/crash; kdumpctl restart
Rebuilt because "Detected change in /etc/kdump.conf"
Most of the credits for this patch goes to Xunlei Pang <xpang@redhat.com>
for suggesting several improvements.
Signed-off-by: Pratyush Anand <panand@redhat.com>
Acked-by: Xunlei Pang <xlpang@redhat.com>
Acked-by: Baoquan He <xlpang@redhat.com>
Acked-by: Dave Young <dyoung@redhat.com>
2016-05-09 08:17:55 +00:00
|
|
|
fi
|
|
|
|
|
2020-03-05 15:28:53 +00:00
|
|
|
_new_mntpoint="$(get_kdump_mntpoint_from_target $_target)"
|
2017-05-04 06:56:56 +00:00
|
|
|
_dracut_args=$(lsinitrd $TARGET_INITRD -f usr/lib/dracut/build-parameter.txt)
|
kdumpctl: force rebuild in case of file system is modified
kdumpctl passes --device argument if dump target is a raw device. It passes
--mount argument if dump target is either mounted as nfs or as a bulk
device. When dump target device is a root device then it does not pass any
of the above two arguments.
After kdumpctl restart, if there is any change in file system which needs
different dracut arguments, then initramfs must be rebuild.
Modification in filesystem for a raw target does not affect dracut
arguments. So, we do not consider to check any modification if raw target
was specified in kdump.conf.
We might need to change dracut arguments if there is some changes in nfs
and ssh target related files. However, we do not consider them in this
patch.
We mainly consider changes in bulk target specified in kdump.conf. We also
consider changes in bulk and nfs file system, if there was no dump target
specified in kdump.conf but dump path is mounting such file systems.
So the initramfs must be rebuild if, either dump target's persistent path
or it's mount point or its file system type changes. If there is no dump
target specified then, both dump path and root path must mount same device,
otherwise rebuild should be triggered.
Some of the examples when we can need a rebuild:
-- "dump target" is specified as one of ext[234], xfs or btrfs. But after
kdump initramfs building its UUID is changed by reformatting.
-- "dump target" is specified as file system type fs1 (say ext3). But after
kdump initramfs building, user change it to fs2 (say ext4), probably by
a mkfs.ext4 executing on the target device.
-- "dump target" is not specified, but "dump path" mounts a device which
is different than device for root path and either UUID or file system type
is modified after kdump initramfs build.
-- "dump target" is not specified, but "dump path" mounts a nfs device and
nfs host path changes after kdump initramfs build.
Some testing:
Initial conditions:
-- No dump target specified
-- dump path (/var/crash) and root(/) are on same device
-- kdumpctl was already executed once after last modification in
/etc/kdump.conf
# kdumpctl restart
No rebuild
# mkfs.ext2 /dev/md0;mount /dev/md0 /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# mount /dev/md0 /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.ext4 /dev/md0;
# mount /dev/md0 /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.ext4 /dev/mapper/fedora-swap
# mount /dev/mapper/fedora-swap /var/crash/
# kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.btrfs /dev/mapper/fedora-swap -f
# mount /dev/mapper/fedora-swap /var/crash/
# kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# mount /dev/mapper/fedora-swap /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.minix /dev/md0
# mount /dev/md0 /var/crash/; kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# mount 192.168.1.16:/nfsroot /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# mount 192.168.1.16:/nfsroot /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash;mount 192.168.1.12:/nfsroot /var/crash/
# kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
Added "raw /dev/md0" in /etc/kdump.conf
# kdumpctl restart
Rebuilt because "Detected change in /etc/kdump.conf"
# mkfs.ext4 /dev/md0 ;kdumpctl restart
No rebuild
Added "ext4 /dev/md0" in /etc/kdump.conf
# mkfs.ext4 /dev/md0;mount /dev/md0 /mnt
# mkdir /mnt/var;mkdir /mnt/var/crash; kdumpctl restart
Rebuilt because "Detected change in /etc/kdump.conf"
# umount /mnt;mkfs.ext4 /dev/md0;mount /dev/md0 /mnt
# mkdir /mnt/var;mkdir /mnt/var/crash; kdumpctl restart
Rebuilt because "Detected change in /etc/kdump.conf"
Most of the credits for this patch goes to Xunlei Pang <xpang@redhat.com>
for suggesting several improvements.
Signed-off-by: Pratyush Anand <panand@redhat.com>
Acked-by: Xunlei Pang <xlpang@redhat.com>
Acked-by: Baoquan He <xlpang@redhat.com>
Acked-by: Dave Young <dyoung@redhat.com>
2016-05-09 08:17:55 +00:00
|
|
|
if [[ -z "$_dracut_args" ]];then
|
|
|
|
echo "Warning: No dracut arguments found in initrd"
|
|
|
|
return 0
|
|
|
|
fi
|
|
|
|
|
|
|
|
# if --mount argument present then match old and new target, mount
|
|
|
|
# point and file system. If any of them mismatches then rebuild
|
|
|
|
echo $_dracut_args | grep "\-\-mount" &> /dev/null
|
|
|
|
if [[ $? -eq 0 ]];then
|
|
|
|
set -- $(echo $_dracut_args | awk -F "--mount '" '{print $2}' | cut -d' ' -f1,2,3)
|
|
|
|
_old_dev=$1
|
|
|
|
_old_mntpoint=$2
|
|
|
|
_old_fstype=$3
|
|
|
|
[[ $_new_dev = $_old_dev && $_new_mntpoint = $_old_mntpoint && $_new_fstype = $_old_fstype ]] && return 0
|
|
|
|
# otherwise rebuild if target device is not a root device
|
|
|
|
else
|
|
|
|
[[ "$_target" = "$(get_root_fs_device)" ]] && return 0
|
|
|
|
fi
|
|
|
|
|
|
|
|
echo "Detected change in File System"
|
|
|
|
return 1
|
|
|
|
}
|
|
|
|
|
2016-07-20 18:12:21 +00:00
|
|
|
check_wdt_modified()
|
|
|
|
{
|
2017-05-04 06:56:55 +00:00
|
|
|
local -A _drivers
|
|
|
|
local _alldrivers _active _wdtdrv _wdtppath _dir
|
|
|
|
local wd_old wd_new
|
|
|
|
|
2016-07-20 18:12:21 +00:00
|
|
|
is_wdt_mod_omitted
|
|
|
|
[[ $? -eq 0 ]] && return 0
|
|
|
|
[[ -d /sys/class/watchdog/ ]] || return 0
|
|
|
|
|
2017-05-04 06:56:55 +00:00
|
|
|
# Copied logic from dracut 04watchdog/module-setup.sh::installkernel()
|
|
|
|
for _dir in /sys/class/watchdog/*; do
|
|
|
|
[[ -d "$_dir" ]] || continue
|
|
|
|
[[ -f "$_dir/state" ]] || continue
|
|
|
|
_active=$(< "$_dir/state")
|
|
|
|
[[ "$_active" = "active" ]] || continue
|
|
|
|
# device/modalias will return driver of this device
|
|
|
|
_wdtdrv=$(< "$_dir/device/modalias")
|
|
|
|
# There can be more than one module represented by same
|
|
|
|
# modalias. Currently load all of them.
|
|
|
|
# TODO: Need to find a way to avoid any unwanted module
|
|
|
|
# represented by modalias
|
|
|
|
_wdtdrv=$(modprobe --set-version "$kdump_kver" -R $_wdtdrv 2>/dev/null)
|
|
|
|
if [[ $_wdtdrv ]]; then
|
|
|
|
for i in $_wdtdrv; do
|
|
|
|
_drivers[$i]=1
|
|
|
|
done
|
2016-07-20 18:12:21 +00:00
|
|
|
fi
|
2017-05-04 06:56:55 +00:00
|
|
|
# however in some cases, we also need to check that if there is
|
|
|
|
# a specific driver for the parent bus/device. In such cases
|
|
|
|
# we also need to enable driver for parent bus/device.
|
|
|
|
_wdtppath=$(readlink -f "$_dir/device")
|
|
|
|
while [[ -d "$_wdtppath" ]] && [[ "$_wdtppath" != "/sys" ]]; do
|
|
|
|
_wdtppath=$(readlink -f "$_wdtppath/..")
|
|
|
|
[[ -f "$_wdtppath/modalias" ]] || continue
|
|
|
|
|
|
|
|
_wdtdrv=$(< "$_wdtppath/modalias")
|
|
|
|
_wdtdrv=$(modprobe --set-version "$kdump_kver" -R $_wdtdrv 2>/dev/null)
|
|
|
|
if [[ $_wdtdrv ]]; then
|
|
|
|
for i in $_wdtdrv; do
|
|
|
|
_drivers[$i]=1
|
|
|
|
done
|
|
|
|
fi
|
|
|
|
done
|
2016-07-20 18:12:21 +00:00
|
|
|
done
|
|
|
|
|
2017-05-04 06:56:55 +00:00
|
|
|
# ensure that watchdog module is loaded as early as possible
|
|
|
|
_alldrivers="${!_drivers[*]}"
|
|
|
|
[[ $_alldrivers ]] && wd_new="rd.driver.pre=${_alldrivers// /,}"
|
|
|
|
wd_old=$(lsinitrd $TARGET_INITRD -f etc/cmdline.d/00-watchdog.conf)
|
2016-07-20 18:12:21 +00:00
|
|
|
|
2017-05-04 06:56:55 +00:00
|
|
|
[[ "$wd_old" = "$wd_new" ]] && return 0
|
|
|
|
|
|
|
|
return 1
|
2016-07-20 18:12:21 +00:00
|
|
|
}
|
|
|
|
|
2016-05-09 08:17:52 +00:00
|
|
|
# returns 0 if system is not modified
|
|
|
|
# returns 1 if system is modified
|
|
|
|
# returns 2 if system modification is invalid
|
|
|
|
check_system_modified()
|
|
|
|
{
|
2016-05-09 08:17:53 +00:00
|
|
|
local ret
|
|
|
|
|
2016-05-09 08:17:52 +00:00
|
|
|
[[ -f $TARGET_INITRD ]] || return 1
|
|
|
|
|
2016-05-09 08:17:53 +00:00
|
|
|
check_files_modified
|
|
|
|
ret=$?
|
|
|
|
if [ $ret -ne 0 ]; then
|
|
|
|
return $ret
|
|
|
|
fi
|
|
|
|
|
kdumpctl: force rebuild in case of file system is modified
kdumpctl passes --device argument if dump target is a raw device. It passes
--mount argument if dump target is either mounted as nfs or as a bulk
device. When dump target device is a root device then it does not pass any
of the above two arguments.
After kdumpctl restart, if there is any change in file system which needs
different dracut arguments, then initramfs must be rebuild.
Modification in filesystem for a raw target does not affect dracut
arguments. So, we do not consider to check any modification if raw target
was specified in kdump.conf.
We might need to change dracut arguments if there is some changes in nfs
and ssh target related files. However, we do not consider them in this
patch.
We mainly consider changes in bulk target specified in kdump.conf. We also
consider changes in bulk and nfs file system, if there was no dump target
specified in kdump.conf but dump path is mounting such file systems.
So the initramfs must be rebuild if, either dump target's persistent path
or it's mount point or its file system type changes. If there is no dump
target specified then, both dump path and root path must mount same device,
otherwise rebuild should be triggered.
Some of the examples when we can need a rebuild:
-- "dump target" is specified as one of ext[234], xfs or btrfs. But after
kdump initramfs building its UUID is changed by reformatting.
-- "dump target" is specified as file system type fs1 (say ext3). But after
kdump initramfs building, user change it to fs2 (say ext4), probably by
a mkfs.ext4 executing on the target device.
-- "dump target" is not specified, but "dump path" mounts a device which
is different than device for root path and either UUID or file system type
is modified after kdump initramfs build.
-- "dump target" is not specified, but "dump path" mounts a nfs device and
nfs host path changes after kdump initramfs build.
Some testing:
Initial conditions:
-- No dump target specified
-- dump path (/var/crash) and root(/) are on same device
-- kdumpctl was already executed once after last modification in
/etc/kdump.conf
# kdumpctl restart
No rebuild
# mkfs.ext2 /dev/md0;mount /dev/md0 /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# mount /dev/md0 /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.ext4 /dev/md0;
# mount /dev/md0 /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.ext4 /dev/mapper/fedora-swap
# mount /dev/mapper/fedora-swap /var/crash/
# kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.btrfs /dev/mapper/fedora-swap -f
# mount /dev/mapper/fedora-swap /var/crash/
# kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# mount /dev/mapper/fedora-swap /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash;mkfs.minix /dev/md0
# mount /dev/md0 /var/crash/; kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# mount 192.168.1.16:/nfsroot /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# mount 192.168.1.16:/nfsroot /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
# kdumpctl restart
No rebuild
# umount /var/crash;mount 192.168.1.12:/nfsroot /var/crash/
# kdumpctl restart
Rebuilt because "Detected change in File System"
# umount /var/crash/;kdumpctl restart
Rebuilt because "Detected change in File System"
Added "raw /dev/md0" in /etc/kdump.conf
# kdumpctl restart
Rebuilt because "Detected change in /etc/kdump.conf"
# mkfs.ext4 /dev/md0 ;kdumpctl restart
No rebuild
Added "ext4 /dev/md0" in /etc/kdump.conf
# mkfs.ext4 /dev/md0;mount /dev/md0 /mnt
# mkdir /mnt/var;mkdir /mnt/var/crash; kdumpctl restart
Rebuilt because "Detected change in /etc/kdump.conf"
# umount /mnt;mkfs.ext4 /dev/md0;mount /dev/md0 /mnt
# mkdir /mnt/var;mkdir /mnt/var/crash; kdumpctl restart
Rebuilt because "Detected change in /etc/kdump.conf"
Most of the credits for this patch goes to Xunlei Pang <xpang@redhat.com>
for suggesting several improvements.
Signed-off-by: Pratyush Anand <panand@redhat.com>
Acked-by: Xunlei Pang <xlpang@redhat.com>
Acked-by: Baoquan He <xlpang@redhat.com>
Acked-by: Dave Young <dyoung@redhat.com>
2016-05-09 08:17:55 +00:00
|
|
|
check_dump_fs_modified
|
|
|
|
ret=$?
|
|
|
|
if [ $ret -ne 0 ]; then
|
|
|
|
return $ret
|
|
|
|
fi
|
|
|
|
|
2016-07-20 18:12:21 +00:00
|
|
|
check_wdt_modified
|
|
|
|
if [ $? -ne 0 ]; then
|
|
|
|
echo "Detected change in watchdog state"
|
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
|
2016-05-09 08:17:52 +00:00
|
|
|
return 0
|
|
|
|
}
|
|
|
|
|
2014-07-15 06:41:54 +00:00
|
|
|
check_rebuild()
|
2011-07-06 19:25:34 +00:00
|
|
|
{
|
2017-08-31 06:27:00 +00:00
|
|
|
local capture_capable_initrd="1"
|
2012-08-06 14:01:29 +00:00
|
|
|
local _force_rebuild force_rebuild="0"
|
2017-04-12 05:44:18 +00:00
|
|
|
local _force_no_rebuild force_no_rebuild="0"
|
2016-05-09 08:17:52 +00:00
|
|
|
local ret system_modified="0"
|
2012-06-14 01:57:27 +00:00
|
|
|
|
2016-11-03 18:46:31 +00:00
|
|
|
setup_initrd
|
2018-10-23 14:13:28 +00:00
|
|
|
|
kdump: Rebuild default initrd for firmware assisted dump
The current kdump infrastructure builds a separate initrd which then
gets loaded into memory by kexec-tools for use by kdump kernel. But
firmware assisted dump (FADUMP) does not use kexec-based approach.
After crash, firmware reboots the partition and loads grub loader
like the normal booting process does. Hence in the FADUMP approach,
the second kernel (after crash) will always use the default initrd
(OS built). So, to support FADUMP, change is required, as in to add
dump capturing steps, in this initrd.
The current kdumpctl script implementation already has the code to
build initrd using mkdumprd. This patch uses the new '--rebuild'
option introduced, in dracut, to incrementally build the initramfs
image. Before rebuilding, we may need to probe the initrd image for
fadump support, to avoid rebuilding the initrd image multiple times
unnecessarily. This can be done using "lsinitrd" tool with the newly
proposed '--mod' option & inspecting the presence of "kdumpbase" in
the list of modules of default initrd image. We rebuild the image if
only "kdumpbase" module is missing in the initrd image. Also, before
rebuilding, a backup of default initrd image is taken.
Kexec-tools package in rhel7 is now enhanced to insert a out-of-tree
kdump module for dracut, which is responsible for adding vmcore
capture steps into initrd, if dracut is invoked with "IN_KDUMP"
environment variable set to 1. mkdumprd script exports "IN_KDUMP=1"
environment variable before invoking dracut to build kdump initrd.
This patch relies on this current mechanism of kdump init script.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
2014-07-24 18:39:07 +00:00
|
|
|
if [ $? -ne 0 ]; then
|
|
|
|
return 1
|
|
|
|
fi
|
2011-07-06 19:25:34 +00:00
|
|
|
|
2017-04-12 05:44:18 +00:00
|
|
|
_force_no_rebuild=`grep ^force_no_rebuild $KDUMP_CONFIG_FILE 2>/dev/null`
|
|
|
|
if [ $? -eq 0 ]; then
|
|
|
|
force_no_rebuild=`echo $_force_no_rebuild | cut -d' ' -f2`
|
|
|
|
if [ "$force_no_rebuild" != "0" ] && [ "$force_no_rebuild" != "1" ];then
|
|
|
|
echo "Error: force_no_rebuild value is invalid"
|
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
fi
|
|
|
|
|
2012-08-06 14:01:29 +00:00
|
|
|
_force_rebuild=`grep ^force_rebuild $KDUMP_CONFIG_FILE 2>/dev/null`
|
|
|
|
if [ $? -eq 0 ]; then
|
|
|
|
force_rebuild=`echo $_force_rebuild | cut -d' ' -f2`
|
|
|
|
if [ "$force_rebuild" != "0" ] && [ "$force_rebuild" != "1" ];then
|
|
|
|
echo "Error: force_rebuild value is invalid"
|
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
fi
|
|
|
|
|
2017-04-12 05:44:18 +00:00
|
|
|
if [[ "$force_no_rebuild" == "1" && "$force_rebuild" == "1" ]]; then
|
|
|
|
echo "Error: force_rebuild and force_no_rebuild are enabled simultaneously in kdump.conf"
|
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
|
|
|
|
# Will not rebuild kdump initrd
|
|
|
|
if [ "$force_no_rebuild" == "1" ]; then
|
|
|
|
return 0
|
|
|
|
fi
|
|
|
|
|
2012-06-14 01:57:27 +00:00
|
|
|
#check to see if dependent files has been modified
|
|
|
|
#since last build of the image file
|
kdump: Rebuild default initrd for firmware assisted dump
The current kdump infrastructure builds a separate initrd which then
gets loaded into memory by kexec-tools for use by kdump kernel. But
firmware assisted dump (FADUMP) does not use kexec-based approach.
After crash, firmware reboots the partition and loads grub loader
like the normal booting process does. Hence in the FADUMP approach,
the second kernel (after crash) will always use the default initrd
(OS built). So, to support FADUMP, change is required, as in to add
dump capturing steps, in this initrd.
The current kdumpctl script implementation already has the code to
build initrd using mkdumprd. This patch uses the new '--rebuild'
option introduced, in dracut, to incrementally build the initramfs
image. Before rebuilding, we may need to probe the initrd image for
fadump support, to avoid rebuilding the initrd image multiple times
unnecessarily. This can be done using "lsinitrd" tool with the newly
proposed '--mod' option & inspecting the presence of "kdumpbase" in
the list of modules of default initrd image. We rebuild the image if
only "kdumpbase" module is missing in the initrd image. Also, before
rebuilding, a backup of default initrd image is taken.
Kexec-tools package in rhel7 is now enhanced to insert a out-of-tree
kdump module for dracut, which is responsible for adding vmcore
capture steps into initrd, if dracut is invoked with "IN_KDUMP"
environment variable set to 1. mkdumprd script exports "IN_KDUMP=1"
environment variable before invoking dracut to build kdump initrd.
This patch relies on this current mechanism of kdump init script.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
2014-07-24 18:39:07 +00:00
|
|
|
if [ -f $TARGET_INITRD ]; then
|
|
|
|
image_time=`stat -c "%Y" $TARGET_INITRD 2>/dev/null`
|
2017-08-31 06:27:00 +00:00
|
|
|
|
|
|
|
#in case of fadump mode, check whether the default/target
|
|
|
|
#initrd is already built with dump capture capability
|
|
|
|
if [ "$DEFAULT_DUMP_MODE" == "fadump" ]; then
|
|
|
|
capture_capable_initrd=$(lsinitrd -f $DRACUT_MODULES_FILE $TARGET_INITRD | grep ^kdumpbase$ | wc -l)
|
|
|
|
fi
|
2011-07-06 19:25:34 +00:00
|
|
|
fi
|
|
|
|
|
2016-05-09 08:17:52 +00:00
|
|
|
check_system_modified
|
|
|
|
ret=$?
|
|
|
|
if [ $ret -eq 2 ]; then
|
|
|
|
return 1
|
|
|
|
elif [ $ret -eq 1 ];then
|
|
|
|
system_modified="1"
|
|
|
|
fi
|
|
|
|
|
2017-05-04 06:56:54 +00:00
|
|
|
if [ $image_time -eq 0 ]; then
|
2012-06-14 01:57:27 +00:00
|
|
|
echo -n "No kdump initial ramdisk found."; echo
|
2017-08-31 06:27:00 +00:00
|
|
|
elif [ "$capture_capable_initrd" == "0" ]; then
|
|
|
|
echo -n "Rebuild $TARGET_INITRD with dump capture support"; echo
|
2012-08-06 14:01:29 +00:00
|
|
|
elif [ "$force_rebuild" != "0" ]; then
|
kdump: Rebuild default initrd for firmware assisted dump
The current kdump infrastructure builds a separate initrd which then
gets loaded into memory by kexec-tools for use by kdump kernel. But
firmware assisted dump (FADUMP) does not use kexec-based approach.
After crash, firmware reboots the partition and loads grub loader
like the normal booting process does. Hence in the FADUMP approach,
the second kernel (after crash) will always use the default initrd
(OS built). So, to support FADUMP, change is required, as in to add
dump capturing steps, in this initrd.
The current kdumpctl script implementation already has the code to
build initrd using mkdumprd. This patch uses the new '--rebuild'
option introduced, in dracut, to incrementally build the initramfs
image. Before rebuilding, we may need to probe the initrd image for
fadump support, to avoid rebuilding the initrd image multiple times
unnecessarily. This can be done using "lsinitrd" tool with the newly
proposed '--mod' option & inspecting the presence of "kdumpbase" in
the list of modules of default initrd image. We rebuild the image if
only "kdumpbase" module is missing in the initrd image. Also, before
rebuilding, a backup of default initrd image is taken.
Kexec-tools package in rhel7 is now enhanced to insert a out-of-tree
kdump module for dracut, which is responsible for adding vmcore
capture steps into initrd, if dracut is invoked with "IN_KDUMP"
environment variable set to 1. mkdumprd script exports "IN_KDUMP=1"
environment variable before invoking dracut to build kdump initrd.
This patch relies on this current mechanism of kdump init script.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
2014-07-24 18:39:07 +00:00
|
|
|
echo -n "Force rebuild $TARGET_INITRD"; echo
|
2016-05-09 08:17:52 +00:00
|
|
|
elif [ "$system_modified" != "0" ]; then
|
|
|
|
:
|
2012-06-14 01:57:27 +00:00
|
|
|
else
|
|
|
|
return 0
|
2011-07-06 19:25:34 +00:00
|
|
|
fi
|
|
|
|
|
kdump: Rebuild default initrd for firmware assisted dump
The current kdump infrastructure builds a separate initrd which then
gets loaded into memory by kexec-tools for use by kdump kernel. But
firmware assisted dump (FADUMP) does not use kexec-based approach.
After crash, firmware reboots the partition and loads grub loader
like the normal booting process does. Hence in the FADUMP approach,
the second kernel (after crash) will always use the default initrd
(OS built). So, to support FADUMP, change is required, as in to add
dump capturing steps, in this initrd.
The current kdumpctl script implementation already has the code to
build initrd using mkdumprd. This patch uses the new '--rebuild'
option introduced, in dracut, to incrementally build the initramfs
image. Before rebuilding, we may need to probe the initrd image for
fadump support, to avoid rebuilding the initrd image multiple times
unnecessarily. This can be done using "lsinitrd" tool with the newly
proposed '--mod' option & inspecting the presence of "kdumpbase" in
the list of modules of default initrd image. We rebuild the image if
only "kdumpbase" module is missing in the initrd image. Also, before
rebuilding, a backup of default initrd image is taken.
Kexec-tools package in rhel7 is now enhanced to insert a out-of-tree
kdump module for dracut, which is responsible for adding vmcore
capture steps into initrd, if dracut is invoked with "IN_KDUMP"
environment variable set to 1. mkdumprd script exports "IN_KDUMP=1"
environment variable before invoking dracut to build kdump initrd.
This patch relies on this current mechanism of kdump init script.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
2014-07-24 18:39:07 +00:00
|
|
|
echo "Rebuilding $TARGET_INITRD"
|
2012-06-14 01:57:27 +00:00
|
|
|
rebuild_initrd
|
|
|
|
return $?
|
2011-07-06 19:25:34 +00:00
|
|
|
}
|
|
|
|
|
2016-11-15 09:07:59 +00:00
|
|
|
# Load the kdump kernel specified in /etc/sysconfig/kdump
|
2011-07-06 19:25:34 +00:00
|
|
|
# If none is specified, try to load a kdump kernel with the same version
|
|
|
|
# as the currently running kernel.
|
2014-07-15 06:41:54 +00:00
|
|
|
load_kdump()
|
2011-07-06 19:25:34 +00:00
|
|
|
{
|
2018-05-22 10:15:07 +00:00
|
|
|
KEXEC_ARGS=$(prepare_kexec_args "${KEXEC_ARGS}")
|
|
|
|
KDUMP_COMMANDLINE=$(prepare_cmdline "${KDUMP_COMMANDLINE}" "${KDUMP_COMMANDLINE_REMOVE}" "${KDUMP_COMMANDLINE_APPEND}")
|
2011-07-06 19:25:34 +00:00
|
|
|
|
kdump-lib: switch to the kexec_file_load() syscall on x86_64 by default
UEFI Secure boot is a signature verification mechanism, designed to
prevent malicious code being loaded and executed at the early boot
stage. This makes sure that code executed is trusted by firmware.
Previously, with kexec_file_load() interface, kernel prevents unsigned
kernel image from being loaded if secure boot is enabled. So kdump will
detect whether secure boot is enabled firstly, then decide which interface
is chosen to execute, kexec_load() or kexec_file_load(). Otherwise unsigned
kernel loading will fail if secure boot enabled, and kexec_file_load() is
entered.
Now, the implementation of kexec_file_load() is adjusted in below commit.
With this change, if CONFIG_KEXEC_SIG_FORCE is not set, unsigned kernel
still has a chance to be allowed to load under some conditions.
commit 99d5cadfde2b ("kexec_file: split KEXEC_VERIFY_SIG into KEXEC_SIG
and KEXEC_SIG_FORCE")
And in the current Fedora, the CONFIG_KEXEC_SIG_FORCE is not set, only the
CONFIG_KEXEC_SIG and CONFIG_BZIMAGE_VERIFY_SIG are set on x86_64 by default.
It's time to spread kexec_file_load() onto all systems of x86_64, including
Secure-boot platforms and legacy platforms. Please refer to the following
form.
.----------------------------------------------------------------------.
| . | signed kernel | unsigned kernel |
| . types |-----------------------|-----------------------|
| . |Secure boot| Legacy |Secure boot| Legacy |
| . |-----------|-----------|-----------|-----------|
| options . | prev| now | prev| now | | | prev| now |
| . |(file|(file|(only|(file| prev| now |(only|(file|
| . |load)|load)|load)|load)| | |load)|load)|
|----------------------|-----|-----|-----|-----|-----|-----|-----|-----|
|KEXEC_SIG=y | | | | | | | | |
|SIG_FORCE is not set |succ |succ |succ |succ | X | X |succ |succ |
|BZIMAGE_VERIFY_SIG=y | | | | | | | | |
|----------------------|-----|-----|-----|-----|-----|-----|-----|-----|
|KEXEC_SIG=y | | | | | | | | |
|SIG_FORCE is not set | | | | | | | | |
|BZIMAGE_VERIFY_SIG is |fail |fail |succ |fail | X | X |succ |fail |
|not set | | | | | | | | |
|----------------------|-----|-----|-----|-----|-----|-----|-----|-----|
|KEXEC_SIG=y | | | | | | | | |
|SIG_FORCE=y |succ |succ |succ |fail | X | X |succ |fail |
|BZIMAGE_VERIFY_SIG=y | | | | | | | | |
|----------------------|-----|-----|-----|-----|-----|-----|-----|-----|
|KEXEC_SIG=y | | | | | | | | |
|SIG_FORCE=y | | | | | | | | |
|BZIMAGE_VERIFY_SIG is |fail |fail |succ |fail | X | X |succ |fail |
|not set | | | | | | | | |
|----------------------|-----|-----|-----|-----|-----|-----|-----|-----|
|KEXEC_SIG is not set | | | | | | | | |
|SIG_FORCE is not set | | | | | | | | |
|BZIMAGE_VERIFY_SIG is |fail |fail |succ |succ | X | X |succ |succ |
|not set | | | | | | | | |
----------------------------------------------------------------------
Note:
[1] The 'X' indicates that the 1st kernel(unsigned) can not boot when the
Secure boot is enabled.
Hence, in this patch, if on x86_64, let's use the kexec_file_load() only.
See if anything wrong happened in this case, in Fedora firstly for the
time being.
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
Acked-by: Kairui Song <kasong@redhat.com>
2020-01-16 05:47:35 +00:00
|
|
|
if [ "$KDUMP_FILE_LOAD" == "on" ]; then
|
|
|
|
echo "Using kexec file based syscall."
|
2014-09-08 15:35:22 +00:00
|
|
|
KEXEC_ARGS="$KEXEC_ARGS -s"
|
|
|
|
fi
|
|
|
|
|
2011-07-06 19:25:34 +00:00
|
|
|
$KEXEC $KEXEC_ARGS $standard_kexec_args \
|
|
|
|
--command-line="$KDUMP_COMMANDLINE" \
|
2014-09-08 15:35:20 +00:00
|
|
|
--initrd=$TARGET_INITRD $kdump_kernel
|
2011-07-06 19:25:34 +00:00
|
|
|
if [ $? == 0 ]; then
|
2013-02-18 09:05:41 +00:00
|
|
|
echo "kexec: loaded kdump kernel"
|
2011-07-06 19:25:34 +00:00
|
|
|
return 0
|
|
|
|
else
|
2013-02-18 09:05:41 +00:00
|
|
|
echo "kexec: failed to load kdump kernel" >&2
|
kdump-lib: switch to the kexec_file_load() syscall on x86_64 by default
UEFI Secure boot is a signature verification mechanism, designed to
prevent malicious code being loaded and executed at the early boot
stage. This makes sure that code executed is trusted by firmware.
Previously, with kexec_file_load() interface, kernel prevents unsigned
kernel image from being loaded if secure boot is enabled. So kdump will
detect whether secure boot is enabled firstly, then decide which interface
is chosen to execute, kexec_load() or kexec_file_load(). Otherwise unsigned
kernel loading will fail if secure boot enabled, and kexec_file_load() is
entered.
Now, the implementation of kexec_file_load() is adjusted in below commit.
With this change, if CONFIG_KEXEC_SIG_FORCE is not set, unsigned kernel
still has a chance to be allowed to load under some conditions.
commit 99d5cadfde2b ("kexec_file: split KEXEC_VERIFY_SIG into KEXEC_SIG
and KEXEC_SIG_FORCE")
And in the current Fedora, the CONFIG_KEXEC_SIG_FORCE is not set, only the
CONFIG_KEXEC_SIG and CONFIG_BZIMAGE_VERIFY_SIG are set on x86_64 by default.
It's time to spread kexec_file_load() onto all systems of x86_64, including
Secure-boot platforms and legacy platforms. Please refer to the following
form.
.----------------------------------------------------------------------.
| . | signed kernel | unsigned kernel |
| . types |-----------------------|-----------------------|
| . |Secure boot| Legacy |Secure boot| Legacy |
| . |-----------|-----------|-----------|-----------|
| options . | prev| now | prev| now | | | prev| now |
| . |(file|(file|(only|(file| prev| now |(only|(file|
| . |load)|load)|load)|load)| | |load)|load)|
|----------------------|-----|-----|-----|-----|-----|-----|-----|-----|
|KEXEC_SIG=y | | | | | | | | |
|SIG_FORCE is not set |succ |succ |succ |succ | X | X |succ |succ |
|BZIMAGE_VERIFY_SIG=y | | | | | | | | |
|----------------------|-----|-----|-----|-----|-----|-----|-----|-----|
|KEXEC_SIG=y | | | | | | | | |
|SIG_FORCE is not set | | | | | | | | |
|BZIMAGE_VERIFY_SIG is |fail |fail |succ |fail | X | X |succ |fail |
|not set | | | | | | | | |
|----------------------|-----|-----|-----|-----|-----|-----|-----|-----|
|KEXEC_SIG=y | | | | | | | | |
|SIG_FORCE=y |succ |succ |succ |fail | X | X |succ |fail |
|BZIMAGE_VERIFY_SIG=y | | | | | | | | |
|----------------------|-----|-----|-----|-----|-----|-----|-----|-----|
|KEXEC_SIG=y | | | | | | | | |
|SIG_FORCE=y | | | | | | | | |
|BZIMAGE_VERIFY_SIG is |fail |fail |succ |fail | X | X |succ |fail |
|not set | | | | | | | | |
|----------------------|-----|-----|-----|-----|-----|-----|-----|-----|
|KEXEC_SIG is not set | | | | | | | | |
|SIG_FORCE is not set | | | | | | | | |
|BZIMAGE_VERIFY_SIG is |fail |fail |succ |succ | X | X |succ |succ |
|not set | | | | | | | | |
----------------------------------------------------------------------
Note:
[1] The 'X' indicates that the 1st kernel(unsigned) can not boot when the
Secure boot is enabled.
Hence, in this patch, if on x86_64, let's use the kexec_file_load() only.
See if anything wrong happened in this case, in Fedora firstly for the
time being.
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
Acked-by: Kairui Song <kasong@redhat.com>
2020-01-16 05:47:35 +00:00
|
|
|
if [ "$KDUMP_FILE_LOAD" == "on" ]; then
|
|
|
|
echo "kexec_file_load() failed, please try kexec_load()" >&2
|
|
|
|
fi
|
2011-07-06 19:25:34 +00:00
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
}
|
|
|
|
|
2014-07-15 06:41:54 +00:00
|
|
|
check_ssh_config()
|
2012-02-22 03:16:09 +00:00
|
|
|
{
|
2017-05-11 05:44:28 +00:00
|
|
|
while read config_opt config_val; do
|
2012-02-22 03:16:09 +00:00
|
|
|
case "$config_opt" in
|
|
|
|
sshkey)
|
2017-05-11 05:44:29 +00:00
|
|
|
# remove inline comments after the end of a directive.
|
2012-02-22 03:16:09 +00:00
|
|
|
if [ -f "$config_val" ]; then
|
|
|
|
# canonicalize the path
|
|
|
|
SSH_KEY_LOCATION=$(/usr/bin/readlink -m $config_val)
|
|
|
|
else
|
|
|
|
echo "WARNING: '$config_val' doesn't exist, using default value '$SSH_KEY_LOCATION'"
|
|
|
|
fi
|
|
|
|
;;
|
|
|
|
path)
|
|
|
|
SAVE_PATH=$config_val
|
|
|
|
;;
|
2012-07-23 07:31:28 +00:00
|
|
|
ssh)
|
2012-02-22 03:16:09 +00:00
|
|
|
DUMP_TARGET=$config_val
|
|
|
|
;;
|
|
|
|
*)
|
|
|
|
;;
|
|
|
|
esac
|
2019-05-17 08:48:46 +00:00
|
|
|
done <<< "$(read_strip_comments $KDUMP_CONFIG_FILE)"
|
2012-02-22 03:16:09 +00:00
|
|
|
|
|
|
|
#make sure they've configured kdump.conf for ssh dumps
|
2012-03-28 08:23:38 +00:00
|
|
|
local SSH_TARGET=`echo -n $DUMP_TARGET | sed -n '/.*@/p'`
|
2012-02-22 03:16:09 +00:00
|
|
|
if [ -z "$SSH_TARGET" ]; then
|
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
return 0
|
|
|
|
}
|
|
|
|
|
2019-08-12 08:07:39 +00:00
|
|
|
# ipv6 host address may takes a long time to be ready.
|
|
|
|
# Instead of checking against ipv6 address, we just check the network reachable
|
|
|
|
# by the return val of 'ssh'
|
|
|
|
check_and_wait_network_ready()
|
|
|
|
{
|
|
|
|
local start_time=$(date +%s)
|
2019-09-24 03:19:12 +00:00
|
|
|
local warn_once=1
|
2019-08-12 08:07:39 +00:00
|
|
|
local cur
|
|
|
|
local diff
|
2019-08-28 07:22:22 +00:00
|
|
|
local retval
|
|
|
|
local errmsg
|
2019-08-12 08:07:39 +00:00
|
|
|
|
|
|
|
while true; do
|
2019-08-28 07:22:22 +00:00
|
|
|
errmsg=$(ssh -i $SSH_KEY_LOCATION -o BatchMode=yes $DUMP_TARGET mkdir -p $SAVE_PATH 2>&1)
|
|
|
|
retval=$?
|
|
|
|
|
2019-08-12 08:07:39 +00:00
|
|
|
# ssh exits with the exit status of the remote command or with 255 if an error occurred
|
2019-08-28 07:22:22 +00:00
|
|
|
if [ $retval -eq 0 ]; then
|
2019-08-12 08:07:39 +00:00
|
|
|
return 0
|
2019-08-28 07:22:22 +00:00
|
|
|
elif [ $retval -ne 255 ]; then
|
|
|
|
echo "Could not create $DUMP_TARGET:$SAVE_PATH, you should check the privilege on server side" >&2
|
2019-08-12 08:07:39 +00:00
|
|
|
return 1
|
|
|
|
fi
|
2019-08-28 07:22:22 +00:00
|
|
|
|
|
|
|
# if server removes the authorized_keys or, no /root/.ssh/kdump_id_rsa
|
2019-10-17 06:01:14 +00:00
|
|
|
echo $errmsg | grep -q "Permission denied\|No such file or directory\|Host key verification failed"
|
2019-08-28 07:22:22 +00:00
|
|
|
if [ $? -eq 0 ]; then
|
|
|
|
echo "Could not create $DUMP_TARGET:$SAVE_PATH, you probably need to run \"kdumpctl propagate\"" >&2
|
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
|
2019-09-24 03:19:12 +00:00
|
|
|
if [ $warn_once -eq 1 ]; then
|
|
|
|
echo "Network dump target is not usable, waiting for it to be ready"
|
|
|
|
warn_once=0
|
|
|
|
fi
|
|
|
|
echo -n .
|
|
|
|
|
2019-08-12 08:07:39 +00:00
|
|
|
cur=$(date +%s)
|
2019-08-28 07:22:22 +00:00
|
|
|
let "diff = $cur - $start_time"
|
2019-08-12 08:07:39 +00:00
|
|
|
# 60s time out
|
2019-08-28 07:22:22 +00:00
|
|
|
if [ $diff -gt 180 ]; then
|
2019-08-12 08:07:39 +00:00
|
|
|
break;
|
|
|
|
fi
|
|
|
|
sleep 1
|
|
|
|
done
|
|
|
|
|
2019-08-28 07:22:22 +00:00
|
|
|
echo "Could not create $DUMP_TARGET:$SAVE_PATH, ipaddr is not ready yet. You should check network connection" >&2
|
2019-08-12 08:07:39 +00:00
|
|
|
return 1
|
|
|
|
}
|
|
|
|
|
2014-07-15 06:41:54 +00:00
|
|
|
check_ssh_target()
|
2012-02-22 03:16:09 +00:00
|
|
|
{
|
2019-08-12 08:07:39 +00:00
|
|
|
check_and_wait_network_ready
|
|
|
|
if [ $? -ne 0 ]; then
|
2013-02-18 09:05:43 +00:00
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
return 0
|
2012-02-22 03:16:09 +00:00
|
|
|
}
|
|
|
|
|
2014-07-15 06:41:54 +00:00
|
|
|
propagate_ssh_key()
|
2011-07-06 19:25:34 +00:00
|
|
|
{
|
2012-02-22 03:16:09 +00:00
|
|
|
check_ssh_config
|
|
|
|
if [ $? -ne 0 ]; then
|
2013-02-18 09:05:41 +00:00
|
|
|
echo "No ssh config specified in $KDUMP_CONFIG_FILE. Can't propagate" >&2
|
2012-02-22 03:16:09 +00:00
|
|
|
exit 1
|
|
|
|
fi
|
|
|
|
|
|
|
|
local KEYFILE=$SSH_KEY_LOCATION
|
2011-07-06 19:25:34 +00:00
|
|
|
local errmsg="Failed to propagate ssh key"
|
|
|
|
|
|
|
|
#Check to see if we already created key, if not, create it.
|
|
|
|
if [ -f $KEYFILE ]; then
|
|
|
|
echo "Using existing keys..."
|
|
|
|
else
|
|
|
|
echo -n "Generating new ssh keys... "
|
2012-02-22 03:16:09 +00:00
|
|
|
/usr/bin/ssh-keygen -t rsa -f $KEYFILE -N "" 2>&1 > /dev/null
|
2011-07-06 19:25:34 +00:00
|
|
|
echo "done."
|
|
|
|
fi
|
|
|
|
|
|
|
|
#now find the target ssh user and server to contact.
|
2012-02-22 03:16:09 +00:00
|
|
|
SSH_USER=`echo $DUMP_TARGET | cut -d\ -f2 | cut -d@ -f1`
|
|
|
|
SSH_SERVER=`echo $DUMP_TARGET | sed -e's/\(.*@\)\(.*$\)/\2/'`
|
2014-07-15 06:41:54 +00:00
|
|
|
|
2011-07-06 19:25:34 +00:00
|
|
|
#now send the found key to the found server
|
2012-02-22 03:16:09 +00:00
|
|
|
ssh-copy-id -i $KEYFILE $SSH_USER@$SSH_SERVER
|
2011-07-06 19:25:34 +00:00
|
|
|
RET=$?
|
|
|
|
if [ $RET == 0 ]; then
|
|
|
|
echo $KEYFILE has been added to ~$SSH_USER/.ssh/authorized_keys on $SSH_SERVER
|
|
|
|
return 0
|
|
|
|
else
|
2013-02-18 09:05:41 +00:00
|
|
|
echo $errmsg, $KEYFILE failed in transfer to $SSH_SERVER >&2
|
2011-07-06 19:25:34 +00:00
|
|
|
exit 1
|
|
|
|
fi
|
|
|
|
}
|
|
|
|
|
2018-05-11 02:02:39 +00:00
|
|
|
show_reserved_mem()
|
|
|
|
{
|
2018-05-24 05:27:02 +00:00
|
|
|
local mem=$(cat /sys/kernel/kexec_crash_size)
|
|
|
|
local mem_mb=$(expr $mem / 1024 / 1024)
|
2018-05-11 02:02:39 +00:00
|
|
|
|
|
|
|
echo "Reserved "$mem_mb"MB memory for crash kernel"
|
|
|
|
}
|
|
|
|
|
2014-07-24 18:38:36 +00:00
|
|
|
check_current_fadump_status()
|
|
|
|
{
|
|
|
|
# Check if firmware-assisted dump has been registered.
|
|
|
|
rc=`cat $FADUMP_REGISTER_SYS_NODE`
|
|
|
|
[ $rc -eq 1 ] && return 0
|
|
|
|
return 1
|
|
|
|
}
|
|
|
|
|
|
|
|
check_current_status()
|
|
|
|
{
|
|
|
|
if [ $DEFAULT_DUMP_MODE == "fadump" ]; then
|
|
|
|
check_current_fadump_status
|
|
|
|
else
|
|
|
|
check_current_kdump_status
|
|
|
|
fi
|
|
|
|
|
|
|
|
return $?
|
|
|
|
}
|
|
|
|
|
2014-07-15 06:41:54 +00:00
|
|
|
save_raw()
|
2012-04-28 10:01:18 +00:00
|
|
|
{
|
|
|
|
local kdump_dir
|
|
|
|
local raw_target
|
|
|
|
|
|
|
|
raw_target=$(awk '$1 ~ /^raw$/ { print $2; }' $KDUMP_CONFIG_FILE)
|
|
|
|
[ -z "$raw_target" ] && return 0
|
|
|
|
[ -b "$raw_target" ] || {
|
|
|
|
echo "raw partition $raw_target not found"
|
|
|
|
return 1
|
|
|
|
}
|
2018-08-15 13:14:15 +00:00
|
|
|
check_fs=$(lsblk --nodeps -npo FSTYPE $raw_target)
|
|
|
|
if [[ $(echo $check_fs | wc -w) -ne 0 ]]; then
|
|
|
|
echo "Warning: Detected '$check_fs' signature on $raw_target, data loss is expected."
|
|
|
|
return 0
|
|
|
|
fi
|
2012-04-28 10:01:18 +00:00
|
|
|
kdump_dir=`grep ^path $KDUMP_CONFIG_FILE | cut -d' ' -f2-`
|
|
|
|
if [ -z "${kdump_dir}" ]; then
|
|
|
|
coredir="/var/crash/`date +"%Y-%m-%d-%H:%M"`"
|
|
|
|
else
|
|
|
|
coredir="${kdump_dir}/`date +"%Y-%m-%d-%H:%M"`"
|
|
|
|
fi
|
|
|
|
|
|
|
|
mkdir -p "$coredir"
|
|
|
|
[ -d "$coredir" ] || {
|
|
|
|
echo "failed to create $coredir"
|
|
|
|
return 1
|
|
|
|
}
|
|
|
|
if makedumpfile -R $coredir/vmcore <$raw_target >/dev/null 2>&1; then
|
|
|
|
# dump found
|
|
|
|
echo "Dump saved to $coredir/vmcore"
|
|
|
|
# wipe makedumpfile header
|
|
|
|
dd if=/dev/zero of=$raw_target bs=1b count=1 2>/dev/null
|
|
|
|
else
|
|
|
|
rm -rf "$coredir"
|
|
|
|
fi
|
|
|
|
|
|
|
|
return 0
|
|
|
|
}
|
|
|
|
|
2013-06-08 06:22:31 +00:00
|
|
|
local_fs_dump_target()
|
|
|
|
{
|
|
|
|
local _target
|
|
|
|
|
|
|
|
_target=$(egrep "^ext[234]|^xfs|^btrfs|^minix" /etc/kdump.conf)
|
|
|
|
if [ $? -eq 0 ]; then
|
|
|
|
echo $_target|awk '{print $2}'
|
|
|
|
fi
|
|
|
|
}
|
|
|
|
|
2014-07-15 06:41:54 +00:00
|
|
|
path_to_be_relabeled()
|
|
|
|
{
|
2013-06-08 06:22:31 +00:00
|
|
|
local _path _target _mnt="/" _rmnt
|
|
|
|
|
2020-03-05 07:34:06 +00:00
|
|
|
if is_user_configured_dump_target; then
|
|
|
|
if is_mount_in_dracut_args; then
|
|
|
|
return;
|
|
|
|
fi
|
|
|
|
|
2013-06-08 06:22:31 +00:00
|
|
|
_target=$(local_fs_dump_target)
|
|
|
|
if [[ -n "$_target" ]]; then
|
2020-03-12 12:57:08 +00:00
|
|
|
_mnt=$(get_mntpoint_from_target $_target)
|
|
|
|
if ! is_mounted "$_mnt"; then
|
2013-06-08 06:22:31 +00:00
|
|
|
return
|
|
|
|
fi
|
|
|
|
else
|
|
|
|
return
|
|
|
|
fi
|
|
|
|
fi
|
|
|
|
|
|
|
|
_path=$(get_save_path)
|
|
|
|
# if $_path is masked by other mount, we will not relabel it.
|
|
|
|
_rmnt=$(df $_mnt/$_path 2>/dev/null | tail -1 | awk '{ print $NF }')
|
|
|
|
if [ "$_rmnt" == "$_mnt" ]; then
|
|
|
|
echo $_mnt/$_path
|
|
|
|
fi
|
|
|
|
}
|
|
|
|
|
|
|
|
selinux_relabel()
|
|
|
|
{
|
|
|
|
local _path _i _attr
|
|
|
|
|
|
|
|
_path=$(path_to_be_relabeled)
|
|
|
|
if [ -z "$_path" ] || ! [ -d "$_path" ] ; then
|
|
|
|
return
|
|
|
|
fi
|
|
|
|
|
|
|
|
for _i in $(find $_path); do
|
|
|
|
_attr=$(getfattr -m "security.selinux" $_i 2>/dev/null)
|
|
|
|
if [ -z "$_attr" ]; then
|
|
|
|
restorecon $_i;
|
|
|
|
fi
|
|
|
|
done
|
|
|
|
}
|
|
|
|
|
2014-07-15 06:41:54 +00:00
|
|
|
check_fence_kdump_config()
|
2014-04-02 08:33:47 +00:00
|
|
|
{
|
|
|
|
local hostname=`hostname`
|
2017-05-17 03:16:22 +00:00
|
|
|
local ipaddrs=`hostname -I`
|
2014-04-02 08:33:47 +00:00
|
|
|
local nodes=$(get_option_value "fence_kdump_nodes")
|
|
|
|
|
|
|
|
for node in $nodes; do
|
|
|
|
if [ "$node" = "$hostname" ]; then
|
|
|
|
echo "Option fence_kdump_nodes cannot contain $hostname"
|
|
|
|
return 1
|
|
|
|
fi
|
2017-05-17 03:16:22 +00:00
|
|
|
# node can be ipaddr
|
|
|
|
echo $ipaddrs | grep $node > /dev/null
|
|
|
|
if [ $? -eq 0 ]; then
|
|
|
|
echo "Option fence_kdump_nodes cannot contain $node"
|
|
|
|
return 1
|
|
|
|
fi
|
2014-04-02 08:33:47 +00:00
|
|
|
done
|
|
|
|
|
|
|
|
return 0
|
|
|
|
}
|
|
|
|
|
2014-07-24 18:38:36 +00:00
|
|
|
check_dump_feasibility()
|
|
|
|
{
|
|
|
|
if [ $DEFAULT_DUMP_MODE == "fadump" ]; then
|
|
|
|
return 0
|
|
|
|
fi
|
|
|
|
|
|
|
|
check_kdump_feasibility
|
|
|
|
return $?
|
|
|
|
}
|
|
|
|
|
2014-07-24 18:38:53 +00:00
|
|
|
start_fadump()
|
|
|
|
{
|
|
|
|
echo 1 > $FADUMP_REGISTER_SYS_NODE
|
|
|
|
if ! check_current_fadump_status; then
|
|
|
|
echo "fadump: failed to register"
|
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
|
|
|
|
echo "fadump: registered successfully"
|
|
|
|
return 0
|
|
|
|
}
|
|
|
|
|
|
|
|
start_dump()
|
|
|
|
{
|
|
|
|
if [ $DEFAULT_DUMP_MODE == "fadump" ]; then
|
|
|
|
start_fadump
|
|
|
|
else
|
|
|
|
load_kdump
|
|
|
|
fi
|
|
|
|
|
|
|
|
return $?
|
|
|
|
}
|
|
|
|
|
2019-01-17 20:31:23 +00:00
|
|
|
check_failure_action_config()
|
2015-06-11 08:22:11 +00:00
|
|
|
{
|
|
|
|
local default_option
|
2019-01-17 20:31:23 +00:00
|
|
|
local failure_action
|
|
|
|
local option="failure_action"
|
2015-06-11 08:22:11 +00:00
|
|
|
|
|
|
|
default_option=$(awk '$1 ~ /^default$/ {print $2;}' $KDUMP_CONFIG_FILE)
|
2019-01-17 20:31:23 +00:00
|
|
|
failure_action=$(awk '$1 ~ /^failure_action$/ {print $2;}' $KDUMP_CONFIG_FILE)
|
|
|
|
|
|
|
|
if [ -z "$failure_action" -a -z "$default_option" ]; then
|
2015-06-11 08:22:11 +00:00
|
|
|
return 0
|
2019-01-17 20:31:23 +00:00
|
|
|
elif [ -n "$failure_action" -a -n "$default_option" ]; then
|
|
|
|
echo "Cannot specify 'failure_action' and 'default' option together"
|
|
|
|
return 1
|
2015-06-11 08:22:11 +00:00
|
|
|
fi
|
2019-01-17 20:31:23 +00:00
|
|
|
|
|
|
|
if [ -n "$default_option" ]; then
|
|
|
|
option="default"
|
|
|
|
failure_action="$default_option"
|
|
|
|
fi
|
|
|
|
|
|
|
|
case "$failure_action" in
|
|
|
|
reboot|halt|poweroff|shell|dump_to_rootfs)
|
|
|
|
return 0
|
|
|
|
;;
|
|
|
|
*)
|
|
|
|
echo $"Usage kdump.conf: $option {reboot|halt|poweroff|shell|dump_to_rootfs}"
|
|
|
|
return 1
|
|
|
|
esac
|
2015-06-11 08:22:11 +00:00
|
|
|
}
|
|
|
|
|
2019-01-17 20:31:24 +00:00
|
|
|
check_final_action_config()
|
|
|
|
{
|
|
|
|
local final_action
|
|
|
|
|
|
|
|
final_action=$(awk '$1 ~ /^final_action$/ {print $2;}' $KDUMP_CONFIG_FILE)
|
|
|
|
if [ -z "$final_action" ]; then
|
|
|
|
return 0
|
|
|
|
else
|
|
|
|
case "$final_action" in
|
|
|
|
reboot|halt|poweroff)
|
|
|
|
return 0
|
|
|
|
;;
|
|
|
|
*)
|
|
|
|
echo $"Usage kdump.conf: final_action {reboot|halt|poweroff}"
|
|
|
|
return 1
|
|
|
|
esac
|
|
|
|
fi
|
|
|
|
}
|
|
|
|
|
2014-07-15 06:41:54 +00:00
|
|
|
start()
|
2011-07-06 19:25:34 +00:00
|
|
|
{
|
2017-04-29 01:15:00 +00:00
|
|
|
check_dump_feasibility
|
2013-03-11 09:31:25 +00:00
|
|
|
if [ $? -ne 0 ]; then
|
|
|
|
echo "Starting kdump: [FAILED]"
|
2012-04-28 07:17:13 +00:00
|
|
|
return 1
|
2013-03-11 09:31:25 +00:00
|
|
|
fi
|
2012-04-28 07:17:13 +00:00
|
|
|
|
2017-04-29 01:15:00 +00:00
|
|
|
check_config
|
2012-04-28 10:01:18 +00:00
|
|
|
if [ $? -ne 0 ]; then
|
2013-02-18 09:05:42 +00:00
|
|
|
echo "Starting kdump: [FAILED]"
|
2012-04-28 10:01:18 +00:00
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
|
2017-04-29 01:15:00 +00:00
|
|
|
if sestatus 2>/dev/null | grep -q "SELinux status.*enabled"; then
|
|
|
|
selinux_relabel
|
|
|
|
fi
|
|
|
|
|
|
|
|
save_raw
|
2014-02-12 02:31:41 +00:00
|
|
|
if [ $? -ne 0 ]; then
|
|
|
|
echo "Starting kdump: [FAILED]"
|
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
|
2014-07-24 18:38:36 +00:00
|
|
|
check_current_status
|
2014-02-13 03:23:58 +00:00
|
|
|
if [ $? == 0 ]; then
|
|
|
|
echo "Kdump already running: [WARNING]"
|
|
|
|
return 0
|
2011-07-06 19:25:34 +00:00
|
|
|
fi
|
2012-02-22 03:16:09 +00:00
|
|
|
|
2013-02-18 09:05:43 +00:00
|
|
|
if check_ssh_config; then
|
|
|
|
if ! check_ssh_target; then
|
|
|
|
echo "Starting kdump: [FAILED]"
|
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
fi
|
|
|
|
|
2013-03-05 08:07:37 +00:00
|
|
|
check_rebuild
|
2011-07-06 19:25:34 +00:00
|
|
|
if [ $? != 0 ]; then
|
2013-02-18 09:05:42 +00:00
|
|
|
echo "Starting kdump: [FAILED]"
|
2011-07-06 19:25:34 +00:00
|
|
|
return 1
|
|
|
|
fi
|
2014-07-24 18:38:53 +00:00
|
|
|
|
|
|
|
start_dump
|
2011-07-06 19:25:34 +00:00
|
|
|
if [ $? != 0 ]; then
|
2013-02-18 09:05:42 +00:00
|
|
|
echo "Starting kdump: [FAILED]"
|
2011-07-06 19:25:34 +00:00
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
|
2013-02-18 09:05:42 +00:00
|
|
|
echo "Starting kdump: [OK]"
|
2011-07-06 19:25:34 +00:00
|
|
|
}
|
|
|
|
|
2018-10-23 14:13:28 +00:00
|
|
|
reload()
|
|
|
|
{
|
|
|
|
check_current_status
|
|
|
|
if [ $? -ne 0 ]; then
|
2019-10-30 11:32:29 +00:00
|
|
|
echo "Kdump was not running: [WARNING]"
|
2018-10-23 14:13:28 +00:00
|
|
|
fi
|
|
|
|
|
|
|
|
if [ $DEFAULT_DUMP_MODE == "fadump" ]; then
|
2019-02-28 04:50:29 +00:00
|
|
|
reload_fadump
|
|
|
|
return $?
|
2018-10-23 14:13:28 +00:00
|
|
|
else
|
|
|
|
stop_kdump
|
|
|
|
fi
|
|
|
|
|
|
|
|
if [ $? -ne 0 ]; then
|
|
|
|
echo "Stopping kdump: [FAILED]"
|
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
|
|
|
|
echo "Stopping kdump: [OK]"
|
|
|
|
|
|
|
|
setup_initrd
|
|
|
|
if [ $? -ne 0 ]; then
|
|
|
|
echo "Starting kdump: [FAILED]"
|
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
|
|
|
|
start_dump
|
|
|
|
if [ $? -ne 0 ]; then
|
|
|
|
echo "Starting kdump: [FAILED]"
|
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
|
|
|
|
echo "Starting kdump: [OK]"
|
|
|
|
}
|
|
|
|
|
2014-07-24 18:39:00 +00:00
|
|
|
stop_fadump()
|
|
|
|
{
|
|
|
|
echo 0 > $FADUMP_REGISTER_SYS_NODE
|
|
|
|
if check_current_fadump_status; then
|
|
|
|
echo "fadump: failed to unregister"
|
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
|
|
|
|
echo "fadump: unregistered successfully"
|
|
|
|
return 0
|
|
|
|
}
|
|
|
|
|
|
|
|
stop_kdump()
|
2011-07-06 19:25:34 +00:00
|
|
|
{
|
kdump-lib: switch to the kexec_file_load() syscall on x86_64 by default
UEFI Secure boot is a signature verification mechanism, designed to
prevent malicious code being loaded and executed at the early boot
stage. This makes sure that code executed is trusted by firmware.
Previously, with kexec_file_load() interface, kernel prevents unsigned
kernel image from being loaded if secure boot is enabled. So kdump will
detect whether secure boot is enabled firstly, then decide which interface
is chosen to execute, kexec_load() or kexec_file_load(). Otherwise unsigned
kernel loading will fail if secure boot enabled, and kexec_file_load() is
entered.
Now, the implementation of kexec_file_load() is adjusted in below commit.
With this change, if CONFIG_KEXEC_SIG_FORCE is not set, unsigned kernel
still has a chance to be allowed to load under some conditions.
commit 99d5cadfde2b ("kexec_file: split KEXEC_VERIFY_SIG into KEXEC_SIG
and KEXEC_SIG_FORCE")
And in the current Fedora, the CONFIG_KEXEC_SIG_FORCE is not set, only the
CONFIG_KEXEC_SIG and CONFIG_BZIMAGE_VERIFY_SIG are set on x86_64 by default.
It's time to spread kexec_file_load() onto all systems of x86_64, including
Secure-boot platforms and legacy platforms. Please refer to the following
form.
.----------------------------------------------------------------------.
| . | signed kernel | unsigned kernel |
| . types |-----------------------|-----------------------|
| . |Secure boot| Legacy |Secure boot| Legacy |
| . |-----------|-----------|-----------|-----------|
| options . | prev| now | prev| now | | | prev| now |
| . |(file|(file|(only|(file| prev| now |(only|(file|
| . |load)|load)|load)|load)| | |load)|load)|
|----------------------|-----|-----|-----|-----|-----|-----|-----|-----|
|KEXEC_SIG=y | | | | | | | | |
|SIG_FORCE is not set |succ |succ |succ |succ | X | X |succ |succ |
|BZIMAGE_VERIFY_SIG=y | | | | | | | | |
|----------------------|-----|-----|-----|-----|-----|-----|-----|-----|
|KEXEC_SIG=y | | | | | | | | |
|SIG_FORCE is not set | | | | | | | | |
|BZIMAGE_VERIFY_SIG is |fail |fail |succ |fail | X | X |succ |fail |
|not set | | | | | | | | |
|----------------------|-----|-----|-----|-----|-----|-----|-----|-----|
|KEXEC_SIG=y | | | | | | | | |
|SIG_FORCE=y |succ |succ |succ |fail | X | X |succ |fail |
|BZIMAGE_VERIFY_SIG=y | | | | | | | | |
|----------------------|-----|-----|-----|-----|-----|-----|-----|-----|
|KEXEC_SIG=y | | | | | | | | |
|SIG_FORCE=y | | | | | | | | |
|BZIMAGE_VERIFY_SIG is |fail |fail |succ |fail | X | X |succ |fail |
|not set | | | | | | | | |
|----------------------|-----|-----|-----|-----|-----|-----|-----|-----|
|KEXEC_SIG is not set | | | | | | | | |
|SIG_FORCE is not set | | | | | | | | |
|BZIMAGE_VERIFY_SIG is |fail |fail |succ |succ | X | X |succ |succ |
|not set | | | | | | | | |
----------------------------------------------------------------------
Note:
[1] The 'X' indicates that the 1st kernel(unsigned) can not boot when the
Secure boot is enabled.
Hence, in this patch, if on x86_64, let's use the kexec_file_load() only.
See if anything wrong happened in this case, in Fedora firstly for the
time being.
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
Acked-by: Kairui Song <kasong@redhat.com>
2020-01-16 05:47:35 +00:00
|
|
|
if [ "$KDUMP_FILE_LOAD" == "on" ]; then
|
2014-09-08 15:35:21 +00:00
|
|
|
$KEXEC -s -p -u
|
|
|
|
else
|
|
|
|
$KEXEC -p -u
|
|
|
|
fi
|
|
|
|
|
2014-07-24 18:39:00 +00:00
|
|
|
if [ $? != 0 ]; then
|
2014-07-15 06:41:54 +00:00
|
|
|
echo "kexec: failed to unload kdump kernel"
|
2014-07-24 18:39:00 +00:00
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
|
|
|
|
echo "kexec: unloaded kdump kernel"
|
|
|
|
return 0
|
|
|
|
}
|
|
|
|
|
2019-02-28 04:50:29 +00:00
|
|
|
reload_fadump()
|
|
|
|
{
|
|
|
|
echo 1 > $FADUMP_REGISTER_SYS_NODE
|
|
|
|
if [ $? == 0 ]; then
|
|
|
|
echo "fadump: re-registered successfully"
|
|
|
|
return 0
|
|
|
|
else
|
|
|
|
# FADump could fail on older kernel where re-register
|
|
|
|
# support is not enabled. Try stop/start from userspace
|
|
|
|
# to handle such scenario.
|
|
|
|
stop_fadump
|
|
|
|
if [ $? == 0 ]; then
|
|
|
|
start_fadump
|
|
|
|
return $?
|
|
|
|
fi
|
|
|
|
fi
|
|
|
|
|
|
|
|
return 1
|
|
|
|
}
|
|
|
|
|
2014-07-24 18:39:00 +00:00
|
|
|
stop()
|
|
|
|
{
|
|
|
|
if [ $DEFAULT_DUMP_MODE == "fadump" ]; then
|
|
|
|
stop_fadump
|
|
|
|
else
|
|
|
|
stop_kdump
|
|
|
|
fi
|
|
|
|
|
|
|
|
if [ $? != 0 ]; then
|
2013-02-18 09:05:42 +00:00
|
|
|
echo "Stopping kdump: [FAILED]"
|
2011-07-06 19:25:34 +00:00
|
|
|
return 1
|
|
|
|
fi
|
2014-07-24 18:39:00 +00:00
|
|
|
|
|
|
|
echo "Stopping kdump: [OK]"
|
|
|
|
return 0
|
2011-07-06 19:25:34 +00:00
|
|
|
}
|
|
|
|
|
2019-03-29 03:29:30 +00:00
|
|
|
rebuild() {
|
2019-05-23 09:34:05 +00:00
|
|
|
check_config
|
|
|
|
if [ $? -ne 0 ]; then
|
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
|
2019-05-27 06:46:49 +00:00
|
|
|
if check_ssh_config; then
|
|
|
|
if ! check_ssh_target; then
|
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
fi
|
|
|
|
|
2019-03-29 03:29:30 +00:00
|
|
|
setup_initrd
|
|
|
|
if [ $? -ne 0 ]; then
|
|
|
|
return 1
|
|
|
|
fi
|
|
|
|
|
|
|
|
echo "Rebuilding $TARGET_INITRD"
|
|
|
|
rebuild_initrd
|
|
|
|
return $?
|
|
|
|
}
|
|
|
|
|
2012-04-28 10:01:18 +00:00
|
|
|
if [ ! -f "$KDUMP_CONFIG_FILE" ]; then
|
2013-02-18 09:05:42 +00:00
|
|
|
echo "Error: No kdump config file found!" >&2
|
2012-04-28 10:01:18 +00:00
|
|
|
exit 1
|
|
|
|
fi
|
|
|
|
|
2013-11-25 16:23:11 +00:00
|
|
|
main ()
|
|
|
|
{
|
2014-07-24 18:38:36 +00:00
|
|
|
# Determine if the dump mode is kdump or fadump
|
|
|
|
determine_dump_mode
|
|
|
|
|
2013-11-25 16:23:11 +00:00
|
|
|
case "$1" in
|
|
|
|
start)
|
|
|
|
if [ -s /proc/vmcore ]; then
|
|
|
|
save_core
|
|
|
|
reboot
|
|
|
|
else
|
|
|
|
start
|
|
|
|
fi
|
|
|
|
;;
|
|
|
|
stop)
|
|
|
|
stop
|
|
|
|
;;
|
|
|
|
status)
|
2011-07-06 19:25:34 +00:00
|
|
|
EXIT_CODE=0
|
2014-07-24 18:38:36 +00:00
|
|
|
check_current_status
|
2013-11-25 16:23:11 +00:00
|
|
|
case "$?" in
|
|
|
|
0)
|
|
|
|
echo "Kdump is operational"
|
|
|
|
EXIT_CODE=0
|
|
|
|
;;
|
|
|
|
1)
|
|
|
|
echo "Kdump is not operational"
|
|
|
|
EXIT_CODE=3
|
|
|
|
;;
|
|
|
|
esac
|
|
|
|
exit $EXIT_CODE
|
2011-07-06 19:25:34 +00:00
|
|
|
;;
|
2018-10-23 14:13:28 +00:00
|
|
|
reload)
|
|
|
|
reload
|
|
|
|
;;
|
2013-11-25 16:23:11 +00:00
|
|
|
restart)
|
|
|
|
stop
|
|
|
|
start
|
|
|
|
;;
|
2019-03-29 03:29:30 +00:00
|
|
|
rebuild)
|
|
|
|
rebuild
|
|
|
|
;;
|
2013-11-25 16:23:11 +00:00
|
|
|
condrestart)
|
2011-07-06 19:25:34 +00:00
|
|
|
;;
|
2013-11-25 16:23:11 +00:00
|
|
|
propagate)
|
|
|
|
propagate_ssh_key
|
2011-07-06 19:25:34 +00:00
|
|
|
;;
|
2018-05-11 02:02:39 +00:00
|
|
|
showmem)
|
|
|
|
show_reserved_mem
|
|
|
|
;;
|
2013-11-25 16:23:11 +00:00
|
|
|
*)
|
2019-03-29 03:29:30 +00:00
|
|
|
echo $"Usage: $0 {start|stop|status|restart|reload|rebuild|propagate|showmem}"
|
2013-11-25 16:23:11 +00:00
|
|
|
exit 1
|
2011-07-06 19:25:34 +00:00
|
|
|
esac
|
2013-11-25 16:23:11 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
# Other kdumpctl instances will block in queue, until this one exits
|
|
|
|
single_instance_lock
|
|
|
|
|
|
|
|
# To avoid fd 9 leaking, we invoke a subshell, close fd 9 and call main.
|
|
|
|
# So that fd isn't leaking when main is invoking a subshell.
|
|
|
|
(exec 9<&-; main $1)
|
2011-07-06 19:25:34 +00:00
|
|
|
|
|
|
|
exit $?
|