kexec-tools/kdump-lib-initramfs.sh

189 lines
4.2 KiB
Bash
Raw Normal View History

#!/bin/sh
#
# The code in this file will be used in initramfs environment, bash may
# not be the default shell. Any code added must be POSIX compliant.
DEFAULT_PATH="/var/crash/"
KDUMP_CONFIG_FILE="/etc/kdump.conf"
FENCE_KDUMP_CONFIG_FILE="/etc/sysconfig/fence_kdump"
FENCE_KDUMP_SEND="/usr/libexec/fence_kdump_send"
LVM_CONF="/etc/lvm/lvm.conf"
# Read kdump config in well formated style
kdump_read_conf()
{
# Following steps are applied in order: strip trailing comment, strip trailing space,
# strip heading space, match non-empty line, remove duplicated spaces between conf name and value
[ -f "$KDUMP_CONFIG_FILE" ] && sed -n -e "s/#.*//;s/\s*$//;s/^\s*//;s/\(\S\+\)\s*\(.*\)/\1 \2/p" $KDUMP_CONFIG_FILE
}
# Retrieves config value defined in kdump.conf
# $1: config name, sed regexp compatible
kdump_get_conf_val()
{
# For lines matching "^\s*$1\s+", remove matched part (config name including space),
# remove tailing comment, space, then store in hold space. Print out the hold buffer on last line.
[ -f "$KDUMP_CONFIG_FILE" ] &&
sed -n -e "/^\s*\($1\)\s\+/{s/^\s*\($1\)\s\+//;s/#.*//;s/\s*$//;h};\${x;p}" $KDUMP_CONFIG_FILE
}
is_mounted()
{
findmnt -k -n "$1" > /dev/null 2>&1
}
# $1: info type
# $2: mount source type
# $3: mount source
# $4: extra args
get_mount_info()
{
__kdump_mnt=$(findmnt -k -n -r -o "$1" "--$2" "$3" $4)
[ -z "$__kdump_mnt" ] && [ -e "/etc/fstab" ] && __kdump_mnt=$(findmnt -s -n -r -o "$1" "--$2" "$3" $4)
echo "$__kdump_mnt"
}
is_ipv6_address()
{
echo "$1" | grep -q ":"
}
is_fs_type_nfs()
{
[ "$1" = "nfs" ] || [ "$1" = "nfs4" ]
}
is_fs_type_virtiofs()
{
[ "$1" = "virtiofs" ]
}
# If $1 contains dracut_args "--mount", return <filesystem type>
get_dracut_args_fstype()
{
echo $1 | grep "\-\-mount" | sed "s/.*--mount .\(.*\)/\1/" | cut -d' ' -f3
}
# If $1 contains dracut_args "--mount", return <device>
get_dracut_args_target()
{
echo $1 | grep "\-\-mount" | sed "s/.*--mount .\(.*\)/\1/" | cut -d' ' -f1
}
get_save_path()
{
__kdump_path=$(kdump_get_conf_val path)
[ -z "$__kdump_path" ] && __kdump_path=$DEFAULT_PATH
# strip the duplicated "/"
echo "$__kdump_path" | tr -s /
}
get_root_fs_device()
{
findmnt -k -f -n -o SOURCE /
}
# Return the current underlying device of a path, ignore bind mounts
get_target_from_path()
{
__kdump_target=$(df "$1" 2> /dev/null | tail -1 | awk '{print $1}')
[ "$__kdump_target" = "/dev/root" ] && [ ! -e /dev/root ] && __kdump_target=$(get_root_fs_device)
echo "$__kdump_target"
}
get_fs_type_from_target()
{
get_mount_info FSTYPE source "$1" -f
}
get_mntpoint_from_target()
{
local _mntpoint
# get the first TARGET when SOURCE doesn't end with ].
# In most cases, a SOURCE ends with ] when fsroot or subvol exists.
_mntpoint=$(get_mount_info TARGET,SOURCE source "$1" | grep -v "\]$" | awk 'NR==1 { print $1 }')
# fallback to the old way when _mntpoint is empty.
[[ -n "$_mntpoint" ]] || _mntpoint=$(get_mount_info TARGET source "$1" -f )
echo $_mntpoint
}
is_ssh_dump_target()
{
kdump_get_conf_val ssh | grep -q @
}
is_raw_dump_target()
{
[ -n "$(kdump_get_conf_val raw)" ]
}
is_virtiofs_dump_target()
{
if [ -n "$(kdump_get_conf_val virtiofs)" ]; then
return 0
fi
if is_fs_type_virtiofs "$(get_dracut_args_fstype "$(kdump_get_conf_val dracut_args)")"; then
return 0
fi
if is_fs_type_virtiofs "$(get_fs_type_from_target "$(get_target_from_path "$(get_save_path)")")"; then
return 0
fi
return 1
}
is_nfs_dump_target()
{
if [ -n "$(kdump_get_conf_val nfs)" ]; then
return 0
fi
if is_fs_type_nfs "$(get_dracut_args_fstype "$(kdump_get_conf_val dracut_args)")"; then
return 0
fi
if is_fs_type_nfs "$(get_fs_type_from_target "$(get_target_from_path "$(get_save_path)")")"; then
return 0
fi
return 1
}
Re-introduce vmcore creation notification to kdump Upstream: fedora Resolves: RHEL-70214 Conflict: Yes, the conflict is the same as the original c9s commit c5aa4609 ("Introduce vmcore creation notification to kdump") 9ec61f6c ("Return the correct exit code of rebuild initrd") Also this patch cherry-picked the ipv6 fixed in [1]. [1]: https://github.com/rhkdump/kdump-utils/pull/60/files commit 24e76222c740def1d03a506652400fe55959e024 Author: Tao Liu <ltao@redhat.com> Date: Fri Nov 29 16:15:18 2024 +1300 Re-introduce vmcore creation notification to kdump Motivation ========== People may forget to recheck to ensure kdump works, which as a result, a possibility of no vmcores generated after a real system crash. It is unexpected for kdump. It is highly recommended people to test kdump after any system modification, such as: a. after kernel patching or whole yum update, as it might break something on which kdump is dependent, maybe due to introduction of any new bug etc. b. after any change at hardware level, maybe storage, networking, firmware upgrading etc. c. after implementing any new application, like which involves 3rd party modules etc. Though these exceed the range of kdump, however a simple vmcore creation status notification is good to have for now. Design ====== Kdump currently will check any relating files/fs/drivers modified before determine if initrd should rebuild when (re)start. A rebuild is an indicator of such modification, and kdump need to be tested. This will clear the vmcore creation status specified in $VMCORE_CREATION_STATUS, and as a result, a notification of vmcore creation test will be outputted. To test kdump, there is an entry for doing that by "kdumpctl test". It will generate a timestamp string as the ID of the current test, along with a "pending" status in $VMCORE_CREATION_STATUS, then a real crash & dump process will be triggered. After system reboot back to normal, a vmcore creation check will start at "kdumpctl (re)start/status", and will report the results as success/fail/manual status to users. To achieve that, program will first check the status in $VMCORE_CREATION_STATUS. If "pending" status if found, which means the test result is undetermined and need a retrive from remote/local dump folder. Then if test id is found in the dump folder and vmcore is complete, then "pending" would be overwritten by "success", which indicates a successful kdump test. If test id is found in the dump folder but vmcore is incomplete, then it is a "fail" kdump test. If no test id is found, then it is a "manual" status, which indicates users should check the test results manually. If $VMCORE_CREATION_STATUS is already success/fail/manual status, it indicates the test result has already been determined, so the program will not access the remote/local dump folder again. This can limite any unnecessary access to dump target, shorten the time consumption. User should check for the root cause of fail/manual status when get reports. $VMCORE_CREATION_STATUS is used for recording the vmcore creation status of the current env. The format is like: <status> kdump_test_id=<timestamp sec>-<timestamp nanosec> e.g: success kdump_test_id=1729823462-938751820 Which means, there has been a successful kdump test at $(date -d "@1729823462") timestamp for the current env. Timestamp nanosec is only meaningful for uniquify id string. Difference ========== Previously there is one commit 88525ebf ("Introduce vmcore creation notification to kdump") merged and addressing the same issue, but implemented differently: The prev one: Save the $VMCORE_CREATION_STATUS to local drive during the 2nd kernel dumping. If vmcore dumping target is different from $VMCORE_CREATION_STATUS's drive, then the latter one need to be mounted in 2nd kernel. This one: Save the $VMCORE_CREATION_STATUS to local drive only in 1nd kernel, that is, the test result is retrived after 2nd kernel dumping. So it doesn't load or mount other drive in 2nd kernel. The advantage: Extra mounting in 2nd kernel will introduce higher risk of failure, as a result, lower the success of vmcore dumping, which is unaccepted. So keep the code for 2nd kernel as simple is preferred. Usage ===== [root@localhost ~]# kdumpctl restart kdump: kexec: unloaded kdump kernel kdump: Stopping kdump: [OK] kdump: kexec: loaded kdump kernel kdump: Starting kdump: [OK] kdump: Notice: No vmcore creation test performed! [root@localhost ~]# kdumpctl status kdump: Kdump is operational kdump: Notice: No vmcore creation test performed! [root@localhost ~]# kdumpctl test [root@localhost ~]# cat /var/lib/kdump/vmcore-creation.status pending kdump_test_id=1729823462-938751820 [root@localhost ~]# kdumpctl status kdump: Kdump is operational kdump: Notice: Last successful vmcore creation on Fri Oct 25 02:31:02 AM UTC 2024 [root@localhost ~]# cat /var/lib/kdump/vmcore-creation.status success kdump_test_id=1729823462-938751820 [root@localhost ~]# kdumpctl restart kdump: kexec: unloaded kdump kernel kdump: Stopping kdump: [OK] kdump: kexec: loaded kdump kernel kdump: Starting kdump: [OK] kdump: Notice: Last successful vmcore creation on Fri Oct 25 02:31:02 AM UTC 2024 Note: the notification for kdumpctl (re)start/status can be disabled by setting VMCORE_CREATION_NOTIFICATION in /etc/sysconfig/kdump. And fadump is NOT supported for this feature. Signed-off-by: Tao Liu <ltao@redhat.com> Signed-off-by: Tao Liu <ltao@redhat.com>
2024-12-05 04:08:38 +00:00
fs_dump_target()
{
kdump_get_conf_val "ext[234]\|xfs\|btrfs\|minix\|virtiofs"
}
is_fs_dump_target()
{
Re-introduce vmcore creation notification to kdump Upstream: fedora Resolves: RHEL-70214 Conflict: Yes, the conflict is the same as the original c9s commit c5aa4609 ("Introduce vmcore creation notification to kdump") 9ec61f6c ("Return the correct exit code of rebuild initrd") Also this patch cherry-picked the ipv6 fixed in [1]. [1]: https://github.com/rhkdump/kdump-utils/pull/60/files commit 24e76222c740def1d03a506652400fe55959e024 Author: Tao Liu <ltao@redhat.com> Date: Fri Nov 29 16:15:18 2024 +1300 Re-introduce vmcore creation notification to kdump Motivation ========== People may forget to recheck to ensure kdump works, which as a result, a possibility of no vmcores generated after a real system crash. It is unexpected for kdump. It is highly recommended people to test kdump after any system modification, such as: a. after kernel patching or whole yum update, as it might break something on which kdump is dependent, maybe due to introduction of any new bug etc. b. after any change at hardware level, maybe storage, networking, firmware upgrading etc. c. after implementing any new application, like which involves 3rd party modules etc. Though these exceed the range of kdump, however a simple vmcore creation status notification is good to have for now. Design ====== Kdump currently will check any relating files/fs/drivers modified before determine if initrd should rebuild when (re)start. A rebuild is an indicator of such modification, and kdump need to be tested. This will clear the vmcore creation status specified in $VMCORE_CREATION_STATUS, and as a result, a notification of vmcore creation test will be outputted. To test kdump, there is an entry for doing that by "kdumpctl test". It will generate a timestamp string as the ID of the current test, along with a "pending" status in $VMCORE_CREATION_STATUS, then a real crash & dump process will be triggered. After system reboot back to normal, a vmcore creation check will start at "kdumpctl (re)start/status", and will report the results as success/fail/manual status to users. To achieve that, program will first check the status in $VMCORE_CREATION_STATUS. If "pending" status if found, which means the test result is undetermined and need a retrive from remote/local dump folder. Then if test id is found in the dump folder and vmcore is complete, then "pending" would be overwritten by "success", which indicates a successful kdump test. If test id is found in the dump folder but vmcore is incomplete, then it is a "fail" kdump test. If no test id is found, then it is a "manual" status, which indicates users should check the test results manually. If $VMCORE_CREATION_STATUS is already success/fail/manual status, it indicates the test result has already been determined, so the program will not access the remote/local dump folder again. This can limite any unnecessary access to dump target, shorten the time consumption. User should check for the root cause of fail/manual status when get reports. $VMCORE_CREATION_STATUS is used for recording the vmcore creation status of the current env. The format is like: <status> kdump_test_id=<timestamp sec>-<timestamp nanosec> e.g: success kdump_test_id=1729823462-938751820 Which means, there has been a successful kdump test at $(date -d "@1729823462") timestamp for the current env. Timestamp nanosec is only meaningful for uniquify id string. Difference ========== Previously there is one commit 88525ebf ("Introduce vmcore creation notification to kdump") merged and addressing the same issue, but implemented differently: The prev one: Save the $VMCORE_CREATION_STATUS to local drive during the 2nd kernel dumping. If vmcore dumping target is different from $VMCORE_CREATION_STATUS's drive, then the latter one need to be mounted in 2nd kernel. This one: Save the $VMCORE_CREATION_STATUS to local drive only in 1nd kernel, that is, the test result is retrived after 2nd kernel dumping. So it doesn't load or mount other drive in 2nd kernel. The advantage: Extra mounting in 2nd kernel will introduce higher risk of failure, as a result, lower the success of vmcore dumping, which is unaccepted. So keep the code for 2nd kernel as simple is preferred. Usage ===== [root@localhost ~]# kdumpctl restart kdump: kexec: unloaded kdump kernel kdump: Stopping kdump: [OK] kdump: kexec: loaded kdump kernel kdump: Starting kdump: [OK] kdump: Notice: No vmcore creation test performed! [root@localhost ~]# kdumpctl status kdump: Kdump is operational kdump: Notice: No vmcore creation test performed! [root@localhost ~]# kdumpctl test [root@localhost ~]# cat /var/lib/kdump/vmcore-creation.status pending kdump_test_id=1729823462-938751820 [root@localhost ~]# kdumpctl status kdump: Kdump is operational kdump: Notice: Last successful vmcore creation on Fri Oct 25 02:31:02 AM UTC 2024 [root@localhost ~]# cat /var/lib/kdump/vmcore-creation.status success kdump_test_id=1729823462-938751820 [root@localhost ~]# kdumpctl restart kdump: kexec: unloaded kdump kernel kdump: Stopping kdump: [OK] kdump: kexec: loaded kdump kernel kdump: Starting kdump: [OK] kdump: Notice: Last successful vmcore creation on Fri Oct 25 02:31:02 AM UTC 2024 Note: the notification for kdumpctl (re)start/status can be disabled by setting VMCORE_CREATION_NOTIFICATION in /etc/sysconfig/kdump. And fadump is NOT supported for this feature. Signed-off-by: Tao Liu <ltao@redhat.com> Signed-off-by: Tao Liu <ltao@redhat.com>
2024-12-05 04:08:38 +00:00
[ -n "$(fs_dump_target)" ]
}
is_lvm2_thinp_device()
{
_device_path=$1
_lvm2_thin_device=$(lvm lvs -S 'lv_layout=sparse && lv_layout=thin' \
--nosuffix --noheadings -o vg_name,lv_name "$_device_path" 2> /dev/null)
[ -n "$_lvm2_thin_device" ]
}
Address the cases where a NIC has a different name in kdump kernel Resolves: bz2076416 Upstream: Fedora Conflict: None commit 568623e69a32266a3b00225b9de6a93435c44474 Author: Coiby Xu <coxu@redhat.com> Date: Thu Sep 23 14:25:01 2021 +0800 Address the cases where a NIC has a different name in kdump kernel A NIC may get a different name in the kdump kernel from 1st kernel in cases like, - kernel assigned network interface names are not persistent e.g. [1] - there is an udev rule to rename the NIC in the 1st kernel but the kdump initrd may not have that rule e.g. [2] If NM tries to match a NIC with a connection profile based on NIC name i.e. connection.interface-name, it will fail the above bases. A simple solution is to ask NM to match a connection profile by MAC address. Note we don't need to do this for user-created NICs like vlan, bridge and bond. An remaining issue is passing the name of a NIC via the kdumpnic dracut command line parameter which requires passing ifname=<interface>:<MAC> to have fixed NIC name. But we can simply drop this requirement. kdumpnic is needed because kdump needs to get the IP by NIC name and use the IP to created a dumping folder named "{IP}-{DATE}". We can simply pass the IP to the kdump kernel directly via a new dracut command line parameter kdumpip instead. In addition to the benefit of simplifying the code, there are other three benefits brought by this approach, - make use of whatever network to transfer the vmcore. Because as long as we have the network to we don't care which NIC is active. - if obtained IP in the kdump kernel is different from the one in the 1st kernel. "{IP}-{DATE}" would better tell where the dumped vmcore comes from. - without passing ifname=<interface>:<MAC> to kdump initrd, the issue of there are two interfaces with the same MAC address for Azure Hyper-V NIC SR-IOV [3] is resolved automatically. [1] https://bugzilla.redhat.com/show_bug.cgi?id=1121778 [2] https://bugzilla.redhat.com/show_bug.cgi?id=810107 [3] https://bugzilla.redhat.com/show_bug.cgi?id=1962421 Signed-off-by: Coiby Xu <coxu@redhat.com> Reviewed-by: Thomas Haller <thaller@redhat.com> Reviewed-by: Philipp Rudo <prudo@redhat.com> Signed-off-by: Coiby Xu <coxu@redhat.com>
2022-11-23 01:42:18 +00:00
kdump_get_ip_route()
{
if ! _route=$(/sbin/ip -o route get to "$1" 2>&1); then
exit 1
fi
echo "$_route"
}
kdump_get_ip_route_field()
{
echo "$1" | sed -n -e "s/^.*\<$2\>\s\+\(\S\+\).*$/\1/p"
}