On 03/29/17 at 03:26pm, Xunlei Pang wrote:
On 03/29/2017 at 03:10 PM, Dave Young wrote:
> Hi Xunlei,
>
> On 03/29/17 at 02:38pm, Xunlei Pang wrote:
>> We replace "reserved_memory= 8192" with "reserved_memory =
1024"
>> in /etc/lvm/lvm.conf used by "lvm2", this can save 7MB peak memory
>> consumption, so we lower the possibility of kdump OOM.
>>
>> For kdump, we don't have too many lvm targets, lvm2 locates in the
>> RAM(rootfs), so don't need that much memory, as discussed with lvm
>> people, they agree that we can use 1MB under kdump as long as there
>> are not so many lvm targets invloved.
>>
>> We modify /etc/lvm/lvm.conf when "99kdumpbase" install() is executed,
>> because it is parsed after "90lvm" by dracut.
>>
>> Signed-off-by: Xunlei Pang <xlpang(a)redhat.com>
>> ---
>> dracut-module-setup.sh | 5 +++++
>> 1 file changed, 5 insertions(+)
>>
>> diff --git a/dracut-module-setup.sh b/dracut-module-setup.sh
>> index 1f96bb8..674f3ee 100755
>> --- a/dracut-module-setup.sh
>> +++ b/dracut-module-setup.sh
>> @@ -743,4 +743,9 @@ install() {
>> # target. Ideally all this should be pushed into dracut iscsi module
>> # at some point of time.
>> kdump_check_iscsi_targets
>> +
>> + # For the lvm type target under kdump, in /etc/lvm/lvm.conf we can safely
replace
>> + # "reserved_memory = 8192" with "reserved_memory =
1024" to lower memory pressure
> The code is setting reserved_memory = 1024 unconditionally, not only for
> relacing 8192 with 1024, need update the commit to match the code. Also
> the line is too long as well.
It only does replacing actually, but if lvm.conf doesn't exist, the code actually
does nothing.
Yes, I got that, what I means is if someone set it as
reserved_memory=4096 we still set it to 1024, so the comment just say
8192 -> 1024. We can just say we will unconditionally set it as 1024.
>> + sed -i -e
's/\(^[[:space:]]*\)reserved_memory[[:space:]]*=[[:space:]]*[[:digit:]]*/ \
>> + \1reserved_memory = 1024/' ${initdir}/etc/lvm/lvm.conf
&>/dev/null
> Nitpick, this can be changed to 80 columns alignment, like move sed -i
> -e in a seperate line.
ok, but it's a little hard for shell, we have many code very long including dracut,
what about the following, I make each independent part one line?
sed -i -e \
's/\(^[[:space:]]*\)reserved_memory[[:space:]]*=[[:space:]]*[[:digit:]]*/\1reserved_memory
= 1024/' \
${initdir}/etc/lvm/lvm.conf &>/dev/null
Yes, it is better, maybe less indent for the last two lines, it will be
within 80 columns
>
> Regards,
> Xunlei
>
> >
> >> }
> >> --
> >> 1.8.3.1
> >> _______________________________________________
> >> kexec mailing list -- kexec(a)lists.fedoraproject.org
> >> To unsubscribe send an email to kexec-leave(a)lists.fedoraproject.org
> > Thanks
> > Dave
> _______________________________________________
> kexec mailing list -- kexec(a)lists.fedoraproject.org
> To unsubscribe send an email to kexec-leave(a)lists.fedoraproject.org