[lustre-discuss] Lustre newbie problems: formatting disk or partition with Lustre filesystem fails

Stefano Turolla turolla at genzentrum.lmu.de
Mon Nov 28 06:58:39 PST 2016


Hi Oliver,

thanks for the quick reply, I am maybe doing the wrong thing. What I am
trying to achieve is to have a Lustre volume to be shared among the
nodes, and the 30TB is the size of existing storage.

Should I create a separate (and much smaller) partition for MDT? How I
then set up the data partition to be shared? For the time being I would
prefer having MGS and MDT on the same system.

thanks

Stefano




On 11/28/2016 03:41 PM, Oliver Mangold wrote:
> Hi Stefano,
>
> FYI, the correct syntax to pass options to mkfs.ext4 is 
> '--mkfsoptions=...', so you might want to try something like 
> "--mkfsoptions=-O extents".
>
> But your numbers also make me a little unsure whether what you try to do 
> is really what you want. 30 TB is a quite impressive size for a MDT for 
> a small cluster. That would be space for 20 billion inodes.
>
> Best regards,
>
> Oliver
>
> On 28.11.2016 14:56, Stefano Turolla wrote:
>> Dear all,
>> I am very new to Lustre and I am pretty sure I am missing something, but
>> here is my problem.
>>
>> I am setting up Lustre for a small Linux cluster ( 6 nodes ).
>>
>> We have a PowerVault MD3800i Storage device connetcted throug iSCSI,
>> disk size is around 30TB
>>
>> The "master" runs scientific linux 7.23 and I installed lustre 2.8 from
>> https://downloads.hpdd.intel.com/public/lustre/latest-feature-release/el7.2.1511/server/RPMS/x86_64/
>>
>> Following rpms have been installed
>> [root at newmaster ~]# rpm -qa|grep lustre
>> perf-3.10.0-327.3.1.el7_lustre.x86_64
>> kernel-3.10.0-327.3.1.el7_lustre.x86_64
>> lustre-modules-2.8.0-3.10.0_327.3.1.el7_lustre.x86_64.x86_64
>> python-perf-3.10.0-327.3.1.el7_lustre.x86_64
>> kernel-headers-3.10.0-327.3.1.el7_lustre.x86_64
>> kernel-tools-libs-3.10.0-327.3.1.el7_lustre.x86_64
>> kernel-tools-libs-devel-3.10.0-327.3.1.el7_lustre.x86_64
>> lustre-osd-ldiskfs-mount-2.8.0-3.10.0_327.3.1.el7_lustre.x86_64.x86_64
>> lustre-osd-ldiskfs-2.8.0-3.10.0_327.3.1.el7_lustre.x86_64.x86_64
>> kernel-devel-3.10.0-327.3.1.el7_lustre.x86_64
>> kernel-tools-3.10.0-327.3.1.el7_lustre.x86_64
>> lustre-2.8.0-3.10.0_327.3.1.el7_lustre.x86_64.x86_64
>>
>> The lnet module is loaded
>> [root at newmaster ~]# cat /etc/modprobe.d/lustre.conf
>> options lnet networks=tcp0(eth2)
>>
>> The problem is the following: when I try to format the disk (entire
>> disk) with lustre ( MDT and MGS are on the same server for the test)  I
>> get the following error:
>>
>> [root at newmaster ~]# mkfs.lustre --reformat --fsname=seq_data --mgs --mdt
>> --index=0 /dev/sdb
>>
>>     Permanent disk data:
>> Target:     seq_data:MDT0000
>> Index:      0
>> Lustre FS:  seq_data
>> Mount type: ldiskfs
>> Flags:      0x65
>>                (MDT MGS first_time update )
>> Persistent mount opts: user_xattr,errors=remount-ro
>> Parameters:
>>
>> device size = 31457280MB
>> formatting backing filesystem ldiskfs on /dev/sdb
>>          target name  seq_data:MDT0000
>>          4k blocks     8053063680
>>          options        -J size=4096 -I 512 -i 2048 -q -O
>> dirdata,uninit_bg,^extents,dir_nlink,quota,huge_file,64bit,flex_bg -E
>> lazy_journal_init -F
>> mkfs_cmd = mke2fs -j -b 4096 -L seq_data:MDT0000  -J size=4096 -I 512 -i
>> 2048 -q -O
>> dirdata,uninit_bg,^extents,dir_nlink,quota,huge_file,64bit,flex_bg -E
>> lazy_journal_init -F /dev/sdb 8053063680
>>     Found a gpt partition table in /dev/sdb
>>     Extents MUST be enabled for a 64-bit filesystem.  Pass -O extents to
>> rectify.
>>
>> mkfs.lustre FATAL: Unable to build fs /dev/sdb (256)
>>
>> mkfs.lustre FATAL: mkfs failed 256
>>
>>
>>
>>
>> If I add the -O extents as suggested I get a different error:
>>
>> [root at newmaster ~]# mkfs.lustre --reformat --fsname=seq_data --mgs --mdt
>> --index=0 -O extents /dev/sdb
>>
>> mkfs.lustre FATAL: OST type is exclusive with MDT,MGS
>> mkfs.lustre: exiting with 22 (Invalid argument)
>>
>>
>>
>> Could you please help me to understand what I am doing wrong?
>> thanks
>> Stefano
>>
>>
> _______________________________________________
> lustre-discuss mailing list
> lustre-discuss at lists.lustre.org
> http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org

-- 
Stefano Turolla
Gene Center
University of Munich
Feodor-Lynen-Str. 25 81377 Munich, Germany
tel. +49 (0)89 2180 71055
email. turolla at genzentrum.lmu.de




More information about the lustre-discuss mailing list