On 2016/05/27 at 14:00, Dave Young wrote:
On 05/26/16 at 09:08pm, Xunlei Pang wrote:
> Update "kexec-kdump-howto" to illustrate the usage of special mount
> information via "dracut_args".
> Suggested-by: Dave Young <dyoung(a)redhat.com>
> Signed-off-by: Xunlei Pang <xlpang(a)redhat.com>
> kexec-kdump-howto.txt | 31 +++++++++++++++++++++++++++++++
> 1 file changed, 31 insertions(+)
> diff --git a/kexec-kdump-howto.txt b/kexec-kdump-howto.txt
> index b4cdc22..f7c6811 100644
> --- a/kexec-kdump-howto.txt
> +++ b/kexec-kdump-howto.txt
> @@ -345,6 +345,37 @@ mount the NFS mount and copy out the vmcore to your NFS server.
> kdump service via '/sbin/systemctl restart kdump.service' to commit this
> to your kdump initrd.
> +Special mount via "dracut_args"
> +Kdump uses dracut to generate initramfs for second kernel. This option
> +allows a user to pass arguments to dracut directly.
> +Users can utilize "dracut_args" to pass "--mount" to kdump,
> +to "man dracut" for the format of "--mount" argument. If there
> +"--mount" specified via "dracut_args", kdump always uses it as
> +mount target without any validation(mounting or checking like mount options,
> +fs size, save path, etc), so users are expected to ensure all the correctness.
> +It conflicts with targets specified via "ext/xfs/btrfs/nfs".
What is the behavior for ssh dump in /etc/kdump.conf being used?
It will fail as well due to "More than one dump targets specified."
Missed ssh, I will add "ssh" here. Thanks for the catching.
> +This is useful when we want to deploy thoundsands of diskless clients using
> +nfs dumping and share the same nfs server. In this scenario, each time the
> +clients are booting up, the kdump initramfs rebuild will be triggered due to
> +the diskless environment, so all the clients will request nfs mount simultaneouly,
> +thereby imposing very much burdon on the shared nfs server. With the help of
> +special mount via "dracut_args", kdump can skip mounting and resolve the
> +Because nfs needn't to be mounted beforehand, some nfs-related ko modules will
> +be explicitly specified as needed together in the "dracut_args"(or via
> +"extra_modules" in /etc/kdump.conf). As an nfs mount example:
> +dracut_args --mount "192.168.1.1:/test /test nfs defaults" --add-driver
For the stateless system use case, it is possible to avoid rebuilding during
every boot. Ie. use a workable initrd with timestamp earlier then /etc/kdump.conf
when admin create the rootfs etc. It is just a guess but there might be ways to
do it. It seems not proper being used in the documentaion.
How about just say that one use case is for nfs dump some admin may not want nfs
server being mounted first.
> +-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 in an absolute path.
> +-There must be double quotation marks behind "--mount", i.e. --mount
> +-Users should do a test first and ensure it works, because kdump does
> + not prepare the mount and check all the validity.
> Remote system via ssh/scp
> Dumping over ssh/scp requires setting up passwordless ssh keys for every
> kexec mailing list