Currently in kdump.sh, we redirect stdout to stderr, because dracut
pre-pivot service (which kdump.sh is running within) only output stderr
to console. That behavior is defined in dracut-pre-pivot.service:
[Service]
...
StandardInput=null
StandardOutput=syslog
StandardError=syslog+console
...
But during testing, it has been observed that systemd will cache stderr
buffer, and first record to syslog (and it's own journal), then copy the
logs to /dev/console. And this practice is somehow unexpected in our
kdump script. We may have suppressed stdout/stderr that hasn't been
write to /dev/console before we run a force reboot.
With this change of redirecting stdout/stderr to /dev/console, kdump.sh
will output everything immediately to console, not cached/hidden by
systemd.
Signed-off-by: WANG Chao <chaowang@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
dump_to_rootfs is a special handling of dump_fs. It's better we merge them
together to cleaup code.
Now dump_fs() function takes two types of $1, a mount point like
/sysroot or a dump target device like /dev/mapper/vg-lv_kdump.
v2: remove -F option in makedumpfile case from Vivek
Signed-off-by: WANG Chao <chaowang@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
After dump the vmcore, explicitly commit changed cache to disk in case
umount fail or chances we'll have an incomplete vmcore.
Signed-off-by: WANG Chao <chaowang@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
When using makedumpfile as core_collector, makedumpfile will show its
own progress bar, it will mix with the monitor_dd_progress and cause confusion.
In this patch just call monitor_dd_progress when core_collector is not
makedumpfile
Signed-off-by: Dave Young <dyoung@redhat.com>
Acked-by: WANG Chao <chaowang@redhat.com>
Currently umount fs happens right after saving vmcore. Therefore vmcore
isn't directly accessible in kdump_post script or shell. This patch moves
the umount fs operation down to the very last part, right before kdump
exits.
The patch adds a global variable MOUNTS to keep track which filesystem
is used. And umount these filesystems at do_default_action() and
do_final_action().
Signed-off-by: WANG Chao <chaowang@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Certain dracut module will mount fs under real root(/sysroot/ or
$NEWROOT/). Thus root fs can not be umounted by `umount /sysroot/`.
We should use `umount -R /sysroot/` to recursively umount root and
its submounts.
v2: do the same for dump_fs() from Baoquan
Signed-off-by: WANG Chao <chaowang@redhat.com>
Signed-off-by: Baoquan He <bhe@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Replace $1/$2 with local variable names in dump_raw() and dump_ssh()
Signed-off-by: WANG Chao <chaowang@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
When makedumpfile failed, it could still generate a invalid vmcore. It's
better to suffix these invalid vmcore files with "-incomplete", as we do
in RHEL6.
Signed-off-by: WANG Chao <chaowang@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
set -x is removed and we'll have little output about the dumping
progress. So it's best to output some messages on the top level to let
user know what's going on.
Signed-off-by: WANG Chao <chaowang@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Currently "set -x" is specified in dracut-kdump.sh and I see the script
execution commands by default on console while testing with F19. That's
not right. This should be done only if user asked for it. Remove it.
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Acked-by: WANG Chao <chaowang@redhat.com>
In dracut-kdump.sh, kdump did not umount rootfs after dump_to_rootfs, just
like dump_fs does. And in kdump, the FINAL_ACTION is "reboot -f", no umount
action is taken.
Even though "sync" has been executed, it's safer to take a "umount rootfs"
action. Anyway no harm to umount.
Signed-off-by: Baoquan He <bhe@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
In kdump.conf, space key is used as delimiter by default.
In kdump_install_conf of dracut-module-setup.sh, if specify
core_collector with a tab delimiter, the tool may not be
copied into kdump-initrd.
E.g, core_collector scp -v
And in dump_ssh of dracut-kdump.sh, dumping will fail caused
by tab key in core_collector.
Here change code to allow tab key as delimiter when specifying
core_collector.
Signed-off-by: Baoquan He <bhe@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
dracut pre-pivot systemd service has below settings:
StandardOutput=syslog
StandardError=syslog+console
Thus kdump_pre/kdump_post output will disapear. Because the output is useful
for users, in case any failure user can watch the console log to see what's
wrong.
Dracut/Systemd people do not want to change the service settings.
So let's redirect the stdout to stderr fo fix it.
Per vivek: redirect whole kdump.sh stdout to stderr instead of only fix
for kdump_pre and kdump_post.
Tested on F19.
Signed-off-by: Dave Young <dyoung@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Save vmcore-dmesg.txt before saving vmcore. For ssh targets, it assumes
that ssh is enabled. No scp logic as I don't have a local copy of
kernel log buffers and saving one will consume extra memory. We
can possibly enhance this logic to save kernel log buffers first locally
and then scp it (For setups which allow scp but disable ssh access).
(log is from Vivek Goyal <vgoyal@redhat.com>)
And add 1 section to describe it in kexec-kdump-howto.txt
v3->v4:
Remove old description of dmesg in kexec-kdump-howto.txt, now
add a new section to describe it, and note user kernel log
buffers won't be available if dump target is raw device.
Signed-off-by: Baoquan He <bhe@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
If CORE_COLLECTOR is makedumpfile, "-F" is only allowed on ssh/raw,
removing it when dump_to_rootfs is necessary.
Signed-off-by: Baoquan He <bhe@redhat.com>
Acked-by: Dave Young <dyoung@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Firstly rename dump_rootfs to dump_to_rootfs to remove the ambiguity
about dump_rootfs. Then add it as one of default options. That means
user can specify dump_to_rootfs to be default action manually, then
it will take action when specified target dump failed.
Secondly, in rhel7 and fedora, when default action is not specified,
the default 'default' is dump_to_rootfs. Namely when specified target
dump failed, the kdump initrd will mount root and save kdump from
initramfs context. However in rhel6, the default 'default' is 'reboot'.
That means when specified target dump failed, the kdump initrd will
reboot systems. For being consistent with rhel6, change the default
'default' back to 'reboot'. And this can also keep logic simple, easier
to understand. Primarily, Our default dump target is root filesystem.
So keeping "default" as "dump_to_rootfs" and trying to dump to root
filesystem again when first attempt fails does not make much sense.
Meanwhile add the relevant description into kdump.conf,kdump.conf.5
and kexec-kdump-howto.txt.
Signed-off-by: Baoquan He <bhe@redhat.com>
Acked-by: Dave Young <dyoung@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
In previous implementation of get_host_ip, global variable HOST_IP
is used to be a intermediate variable. In this case, if finally
failed to get HOST_IP, the original default value is also overwritten.
It's buggy.
Eg. in ssh case, when faied to get host ip, the default local host ip
"127.0.0.1" is lost too. that's not expected.
Change it by adding a local variable as intermediate variable.
Signed-off-by: Baoquan He <bhe@redhat.com>
Acked-by: Dave Young <dyoung@redhat.com>
In previous patch, get_host_ip was introduced to make the dump dir
more readable and back compatible with rhel6. But the implementation
is not good, making variable DATEDIR ambiguous. In this patch, modify
those codes to make it clearer.
About code checking returned value of get_host_ip, if not zero the
function do_default_action is called. That looks buggy, it will
continue going through the later code flow. Add FINAL_ACTION after
invoking do_default_action.
Signed-off-by: Baoquan He <bhe@redhat.com>
Acked-by: Dave Young <dyoung@redhat.com>
After recent dracut network changes, waiting for nic ready stuff is handled
in cmdline hooks when generate udev rules. So wait_for_net_ok is not
necessary any more in kdump.sh, Removing the code here.
Signed-off-by: Dave Young <dyoung@redhat.com>
Acked-by: Baoquan He <bhe@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
To simplify the code logic, remove the previous PIPESTATUS handling because
previous patch "enable pipefail bash option" added pipefail option which will
also resolve this issue.
This is a reverting of below commit:
commit 3111ff74c9
Author: Chao Wang <chaowang@redhat.com>
Date: Mon Aug 6 14:49:52 2012 +0800
Change return value to indicate the result of dump_raw() correctly.
Resolves bz845916.
For raw dump code below:
$CORE_COLLECTOR /proc/vmcore | dd of=$1 bs=$DD_BLKSIZE >> /tmp/dd_progress_f
in case pipe fails it will still return 0, Fix this by return the PIPESTATUS
Signed-off-by: Chao Wang <chaowang@redhat.com>
Acked-by: Dave Young <dyoung@redhat.com>
Signed-off-by: Dave Young <dyoung@redhat.com>
Acked-by: Baoquan He <bhe@redhat.com>
Follwing code will not return 1 if CORE_COLLECTOR /proc/vmcore fail:
$CORE_COLLECTOR /proc/vmcore | ssh $_opt $2 "dd bs=512 of=$_dir/vmcore-incomplete" || return 1
Fix it by enabling pipefail option
Tested in F18 with ssh dump.
Signed-off-by: Dave Young <dyoung@redhat.com>
Acked-by: Baoquan He <bhe@redhat.com>
Resolves: bz880551
Currently on fedora directory name of kdump is like 14.11.12-03:40:30.
It's not as readable as RHEL6. The old format of dir is host_ip-date-time.
This patch changes it back like ole one as below.
192.168.122.234-2012-10-31-05:17:20
If can't get HOST_IP for a network kudmp, we just fail. Becasue most
likely any attempt to save dump will also fail.
Signed-off-by: Baoquan He <bhe@redhat.com>
Acked-by: Dave Young <dyoung@redhat.com>
ssh: add random feeding code
openssh usually depends on /dev/urandom to seeding PRNG (pseudo-random number
generator). To ensure there's sufficient entropy just feed /dev/urandom with
saved /var/lib/random-seed which is saved in installing phase.
dracut_install will print error message, so just exit and do not print error
again.
Todo: other dracut_install fail and exit, such as dd, makedumpfile.
[v2->v3]: use [[:blank:]] for whitespace checking.
Signed-off-by: Dave Young <dyoung@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
We can use not only space but also tab as whitespace,
so s/\ /[[:blank:]] for checking the whitespace
The last commit is intend for checking multiple dump target,
and differentiate ssh and sshkey options. This issue is only
for ssh, so no need to add [[:blank:]] for other dump types to
create a very long code line.
[v1->v2]: use [[:blank:]] instead of [[:space:]]
see expanation in below doc:
http://en.wikipedia.org/wiki/Regular_expression#POSIX_character_classes
[:blank:] [ \t] Space and tab
[:space:] [ \t\r\n\v\f] Whitespace characters
Tested the [:blank:] works well as [:space:]
Signed-off-by: Dave Young <dyoung@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
CC: Cong Wang <amwang@redhat.com>
currently --mount param are retrieved from /proc/mounts, but the device
name could be renamed in initramfs. So here convert them to persistent
names before passing to dracut
lvm canonical dev name is /dev/mapper/lvname-link which will be showed
in /proc/mounts
here fix get_mp function by using findmnt utils to find the mount point.
This patch depends on below dracut patch:
http://permalink.gmane.org/gmane.linux.kernel.initramfs/2903
[chaowang]:
in case device is not mounted we should not echo the mount line in to_mount()
use findmnt -n to strip the header line
for nfs don't pass persistent name to dracut
[vivek]:
improve variable names
Signed-off-by: Dave Young <dyoung@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
When we intall kdump.conf to initramfs we have converted the UUID and LABEL
to device name. So in initramfs these convert code is not used indeed.
Signed-off-by: Dave Young <dyoung@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Resolves bz845916.
For raw dump code below:
$CORE_COLLECTOR /proc/vmcore | dd of=$1 bs=$DD_BLKSIZE >> /tmp/dd_progress_file 2>&1 || return 1
in case pipe fails it will still return 0, Fix this by return the PIPESTATUS[0] at the end of this fuction.
Signed-off-by: Chao Wang <chaowang@redhat.com>
Acked-by: Dave Young <dyoung@redhat.com>
Kdump dracut hooks need to enter emergency shell, currently it directly call
"sh -i -l", with recent dracut this does not work anymore without proper ctty.
change to call dracut lib function _emergency_shell instead of directly call
'sh -i -l'
This patch depends on dracut patch:
http://permalink.gmane.org/gmane.linux.kernel.initramfs/2821
Signed-off-by: Dave Young <dyoung@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Currently net options means either nfs or ssh dump.
Better to split these two into standalone options. That's more clear to user.
after the split, ssh dump need user specify "ssh user@host"
nfs dump need user specify "nfs host:nfsshare"
Signed-off-by: Dave Young <dyoung@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
dump_nfs mount the nfs share to /mnt, it's not what we expected.
Change to use dracut fstab mount infrastructure for this.
After the fix nfs dump is just same as localfs dump.
The --mount for nfs does not work before, it depend on the upstream patch:
1. dracut fix to autoload nfs.ko in case nfs4. the patch post as below:
http://permalink.gmane.org/gmane.linux.kernel.initramfs/2768
Signed-off-by: Dave Young <dyoung@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
nfs will be handled in fstab-sys as well as localfs.
so rename dump_localfs to dump_fs firstly.
Signed-off-by: Dave Young <dyoung@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
in dump_localfs, we should error out if fstab mount fails
Also error out if mkdir fails.
vivek: display error message for debugging purpose.
Signed-off-by: Dave Young <dyoung@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Original patch is from Amerigo, but it has changed a lot:
remove multi dump
refreshed with latest git for the dump instruction function restructure
fixed the behavior of do_kdump_pre, if kdump_pre fails it will reboot
update the docs
check the existance and executable of kdump_pre/post files,
also check the timestamp of them for rebuilding.
refresh patch,
Address comments from vivek:
s/hush/bash in docs
fix the copy-paste error in kdump post error message
s/reboot\/halt/reboot in kexec-kdump-howto.txt
Signed-off-by: Dave Young <dyoung@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
1. only append '-F' for default core_collector internally
2. error out if user does not add '-F' for makedumpfile
v1->v2: refresh because of new core collector verify function
v2->v3: optimize the code, remove local variable for default core_collector
Signed-off-by: Dave Young <dyoung@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
1. add core_collector filter support to ssh dump
2. scp can be specified in kdump.conf
3. error out if no '-F' with makedumpfile
4. add proper explanation and examples to kdump.conf[.5] and kexec-kdump-howto.txt.
v1->v2: add verify_core_collector function
v2->v3: optimize the code, remove local variable for default core_collector
Signed-off-by: Dave Young <dyoung@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
move conf_file as a global variable to the top.
vivek: return earlier if kdump.conf is not found in read_kdump_conf to
avoid the deep indent.
Signed-off-by: Dave Young <dyoung@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
let add_dump_code only deal with real dump instruction,
move other logic out of that function, so it's easier for future features
like kdump_pre and kdump_post.
Signed-off-by: Dave Young <dyoung@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
default action is handled in DUMP_INSTRUCTIONI. If default_action failed
current logic will cause mount_root_run_init, It does not make much sense.
changes to: let reboot -f does not depend on return status of DUMP_INSTRUCTION
Signed-off-by: Dave Young <dyoung@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Do not reboot in dump_rootfs function, also do not directly call dump_rootfs
for default dump. Instead just adding it to DUMP_INSTRUCTION as other
user specified dump target.
Signed-off-by: Dave Young <dyoung@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Resolves: bz805773
simplify logic and remove redundant code:
put do_default_action to add_dump_code
Signed-off-by: Dave Young <dyoung@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Resolves: bz816864
currently kexec-tools maintains kdump dracut modules as below dir hierarchy:
[dave@localhost fedpkg]$ tree kexec-tools/kdump_dracut_modules
kexec-tools/kdump_dracut_modules
`-- 99kdumpbase
|-- kdump.sh
`-- module-setup.sh
But in kexec-tools.spec, sources line is below:
Source100: dracut-files.tbz2
So we need to manually generate the dracut-files.tbz2, upload it, modify
the sources file which contains the md5sum of these binaies, and then do
the building. The more headache problem is this make local tests of
changes to kdump module hard.
To ease future work, move the files to TOPDIR, cp the sources instead.
Thanks for help from dzickus
[v1->v2]: Also remove dracut-files.tbz2 from sources file
ChaoWang: add prefix to dracut files
