On 12/05/13 at 09:51am, Dave Young wrote:
On 12/04/13 at 04:12pm, Vivek Goyal wrote:
> On Wed, Dec 04, 2013 at 07:34:39PM +0800, WANG Chao wrote:
> > Lzo is proven faster than zlib, for large memory machine it will
> > extremely shorten the time for saving vmcore. Let's switch to lzo as the
> > default compression method for makedumpfile.
> >
> > The drawback is lzo has a little less compression ratio than zlib. But
> > considering for most users, speed/time is a more serious concern than
> > vmcore size. So I think default to lzo will benefit most of the users.
> >
> > Signed-off-by: WANG Chao <chaowang(a)redhat.com>
>
> Agreed. On large RAM machines, lzo is net win. People care more about time
> taken to dump as compared to little increase in size of dump. So while
> smaller machines might see little bit increase in size of dump, I guess
> it should still prove to be a good default across the range of machines.
>
> Acked-by: Vivek Goyal <vgoyal(a)redhat.com>
>
> Thanks
> Vivek
>
> > ---
> > dracut-kdump.sh | 2 +-
> > kdump.conf | 6 +++---
> > kexec-kdump-howto.txt | 18 +++++++++---------
> > 3 files changed, 13 insertions(+), 13 deletions(-)
Looks like kdump.conf.5 was missed. The manpage also need update.
We don't have kdump.conf.5. And I don't think we need that.
we don't have kdump.conf.5 in RHEL5/6 and we already detail everything
in kdump.conf and kexec-kdump-howto.txt. So providing a man page for
kdump.conf is a good idea, however we won't benefit much from it.
Thanks
WANG Chao
Thanks for the patch.
> >
> > diff --git a/dracut-kdump.sh b/dracut-kdump.sh
> > index f38891d..14843f3 100755
> > --- a/dracut-kdump.sh
> > +++ b/dracut-kdump.sh
> > @@ -11,7 +11,7 @@ fi
> > set -o pipefail
> > KDUMP_PATH="/var/crash"
> > CORE_COLLECTOR=""
> > -DEFAULT_CORE_COLLECTOR="makedumpfile -c --message-level 1 -d 31"
> > +DEFAULT_CORE_COLLECTOR="makedumpfile -l --message-level 1 -d 31"
> > DMESG_COLLECTOR="/sbin/vmcore-dmesg"
> > DEFAULT_ACTION="reboot -f"
> > DATEDIR=`date +%Y.%m.%d-%T`
> > diff --git a/kdump.conf b/kdump.conf
> > index a0f086a..1bc3064 100644
> > --- a/kdump.conf
> > +++ b/kdump.conf
> > @@ -50,9 +50,9 @@
> > # automatically be populated with a config file
> > # appropriate for the running kernel.
> > # Default core_collector for raw/ssh dump is:
> > -# "makedumpfile -F -c --message-level 1 -d 31".
> > +# "makedumpfile -F -l --message-level 1 -d 31".
> > # Default core_collector for other targets is:
> > -# "makedumpfile -c --message-level 1 -d 31".
> > +# "makedumpfile -l --message-level 1 -d 31".
> > # For core_collector format details please refer to
> > # kexec-kdump-howto.txt or kdump.conf manpage.
> > #
> > @@ -136,7 +136,7 @@
> > #ssh user(a)my.server.com
> > #sshkey /root/.ssh/kdump_id_rsa
> > path /var/crash
> > -#core_collector makedumpfile -c --message-level 1 -d 31
> > +#core_collector makedumpfile -l --message-level 1 -d 31
> > #core_collector scp
> > #kdump_post /var/crash/scripts/kdump-post.sh
> > #kdump_pre /var/crash/scripts/kdump-pre.sh
> > diff --git a/kexec-kdump-howto.txt b/kexec-kdump-howto.txt
> > index d70781b..7ffeab9 100644
> > --- a/kexec-kdump-howto.txt
> > +++ b/kexec-kdump-howto.txt
> > @@ -438,7 +438,7 @@ is a dump filtering and compression utility provided with
kexec-tools. On
> > some architectures, it can drastically reduce the size of your vmcore files,
> > which becomes very useful on systems with large amounts of memory.
> >
> > -A typical setup is 'core_collector makedumpfile -F -c --message-level 1 -d
31',
> > +A typical setup is 'core_collector makedumpfile -F -l --message-level 1 -d
31',
> > but check the output of '/sbin/makedumpfile --help' for a list of all
available
> > options (-i and -g don't need to be specified, they're automatically
taken care
> > of). Note that use of makedumpfile requires that the kernel-debuginfo package
> > @@ -458,11 +458,11 @@ cp --sparse=always /proc/vmcore <dest-path>/vmcore
> >
> > ex2.
> > ---
> > -core_collector "makedumpfile -c --message-level 1 -d 31"
> > +core_collector "makedumpfile -l --message-level 1 -d 31"
> >
> > Above will effectively be translated to:
> >
> > -makedumpfile -c --message-level 1 -d 31 /proc/vmcore <dest-path>/vmcore
> > +makedumpfile -l --message-level 1 -d 31 /proc/vmcore <dest-path>/vmcore
> >
> >
> > For dump targets like raw and ssh, in general, core collector should expect
> > @@ -482,11 +482,11 @@ cat /proc/vmcore | dd of=<target-device>
> >
> > ex4.
> > ---
> > -core_collector "makedumpfile -F -c --message-level 1 -d 31"
> > +core_collector "makedumpfile -F -l --message-level 1 -d 31"
> >
> > Above will effectively be translated to.
> >
> > -makedumpfile -F -c --message-level 1 -d 31 | dd of=<target-device>
> > +makedumpfile -F -l --message-level 1 -d 31 | dd of=<target-device>
> >
> > ssh dumps core_collector examples:
> > ---------
> > @@ -500,11 +500,11 @@ cat /proc/vmcore | ssh <options>
<remote-location> "dd of=path/vmcore"
> >
> > ex6.
> > ---
> > -core_collector "makedumpfile -F -c --message-level 1 -d 31"
> > +core_collector "makedumpfile -F -l --message-level 1 -d 31"
> >
> > Above will effectively be translated to.
> >
> > -makedumpfile -F -c --message-level 1 -d 31 | ssh <options>
<remote-location> "dd of=path/vmcore"
> > +makedumpfile -F -l --message-level 1 -d 31 | ssh <options>
<remote-location> "dd of=path/vmcore"
> >
> > There is one exception to standard output rule for ssh dumps. And that is
> > scp. As scp can handle ssh destinations for file transfers, one can
> > @@ -521,9 +521,9 @@ scp /proc/vmcore <user@host>:path/vmcore
> > About default core collector
> > ----------------------------
> > Default core_collector for ssh/raw dump is:
> > -"makedumpfile -F -c --message-level 1 -d 31".
> > +"makedumpfile -F -l --message-level 1 -d 31".
> > Default core_collector for other targets is:
> > -"makedumpfile -c --message-level 1 -d 31".
> > +"makedumpfile -l --message-level 1 -d 31".
> >
> > Even if core_collector option is commented out in kdump.conf, makedumpfile
> > is default core collector and kdump uses it internally.
> > --
> > 1.8.4.2