* [linux-lvm] lvcreate: adding devices causes an alignment inconsistencies
@ 2020-03-28 16:59 Bernhard Sulzer
2020-04-09 10:24 ` Heinz Mauelshagen
0 siblings, 1 reply; 2+ messages in thread
From: Bernhard Sulzer @ 2020-03-28 16:59 UTC (permalink / raw)
To: linux-lvm
I wanted to create a raid5 array from 3x 7.3TiB drives but whatever I
do, I seemingly can't align the devices in my lvm raid. Am I doing it
wrong, or may there be a bug hiding somewhere?
# sudo vgcreate test --dataalignment 1M /dev/sd{c,d,e}
Volume group "test" successfully created
# sudo lvcreate --type raid5 -L 4T --nosync -n test_data test
Using default stripesize 64.00 KiB.
WARNING: New raid5 won't be synchronised. Don't read what you didn't
write!
Logical volume "test_data" created.
As far as output from lvm commands is concerned, it all looks fine to
me. But why, when looking at lsblk is there so much strange alignment
going on?Is this normal or should I be concerned? (I don't know what
performance I should expect with my setup so I can't say anything about
that). Note, that the same thing happens when I try to create a raid0
# lsblk -t
NAME ALIGNMENT MIN-IO OPT-IO PHY-SEC LOG-SEC ROTA
SCHED RQ-SIZE RA WSAME
sdc 0 4096 0 4096 512 1
mq-deadline 60 128 32M
├─test-test_data_rmeta_0 0 4096 0 4096 512
1 128 128 32M
│ └─test-test_data -1 65536 131072 4096 512
1 128 384 0B
└─test-test_data_rimage_0 0 4096 0 4096 512
1 128 128 32M
└─test-test_data -1 65536 131072 4096 512
1 128 384 0B
sdd 0 4096 0 4096 512 1
mq-deadline 60 128 32M
├─test-test_data_rmeta_1 0 4096 0 4096 512
1 128 128 32M
│ └─test-test_data -1 65536 131072 4096 512
1 128 384 0B
└─test-test_data_rimage_1 0 4096 0 4096 512
1 128 128 32M
└─test-test_data -1 65536 131072 4096 512
1 128 384 0B
sde 0 4096 0 4096 512 1
mq-deadline 60 128 32M
├─test-test_data_rmeta_2 512 4096 0 4096 512
1 128 128 32M
│ └─test-test_data -1 65536 131072 4096 512
1 128 384 0B
└─test-test_data_rimage_2 512 4096 0 4096 512
1 128 128 32M
# dmesg -wH
[ +0.275210] device-mapper: raid: Superblocks created for new raid set
[ +0.002796] md/raid:mdX: device dm-1 operational as raid disk 0
[ +0.000004] md/raid:mdX: device dm-3 operational as raid disk 1
[ +0.000002] md/raid:mdX: device dm-5 operational as raid disk 2
[ +0.000801] md/raid:mdX: raid level 5 active with 3 out of 3 devices,
algorithm 2
[ +0.010185] device-mapper: table: 254:6: adding target device dm-5
caused an alignment inconsistency: physical_block_size=4096,
logical_block_size=512, alignment_offset=512, start=0
[ +0.000005] device-mapper: table: 254:6: adding target device dm-5
caused an alignment inconsistency: physical_block_size=4096,
logical_block_size=512, alignment_offset=512, start=0
[ +0.000084] device-mapper: table: 254:6: adding target device dm-5
caused an alignment inconsistency: physical_block_size=4096,
logical_block_size=512, alignment_offset=512, start=0
[ +0.000003] device-mapper: table: 254:6: adding target device dm-5
caused an alignment inconsistency: physical_block_size=4096,
logical_block_size=512, alignment_offset=512, start=0
[ +0.070102] mdX: bitmap file is out of date, doing full recovery
Platform:
Arch Linux 5.6.0-rc7-1-git-00151-g67d584e33e54 (also tested with Debian
on 4.19, same results)
LVM version: 2.02.186(2) (2019-08-27)
Library version: 1.02.164 (2019-08-27)
Driver version: 4.42.0
Any advice would be much appreciated, thanks
^ permalink raw reply [flat|nested] 2+ messages in thread* Re: [linux-lvm] lvcreate: adding devices causes an alignment inconsistencies
2020-03-28 16:59 [linux-lvm] lvcreate: adding devices causes an alignment inconsistencies Bernhard Sulzer
@ 2020-04-09 10:24 ` Heinz Mauelshagen
0 siblings, 0 replies; 2+ messages in thread
From: Heinz Mauelshagen @ 2020-04-09 10:24 UTC (permalink / raw)
To: LVM general discussion and development, Bernhard Sulzer; +Cc: Mike Snitzer
Bernhard,
the block device stacking limits check failed because of queue alignment
offset 512
when a start at 0 has been configured probably causing performance
issues with your
RaidLV (bdev_stack_limits() in dm-table.c returning error causes the
messages you see).
You won't see any data corruption issues with that kind of error.
Mike,
is this a known issue yet (not seeing a related bz)?
Heinz
On 3/28/20 5:59 PM, Bernhard Sulzer wrote:
> I wanted to create a raid5 array from 3x 7.3TiB drives but whatever I
> do, I seemingly can't align the devices in my lvm raid. Am I doing it
> wrong, or may there be a bug hiding somewhere?
>
> # sudo vgcreate test --dataalignment 1M /dev/sd{c,d,e}
> Volume group "test" successfully created
>
> # sudo lvcreate --type raid5 -L 4T --nosync -n test_data test
> Using default stripesize 64.00 KiB.
> WARNING: New raid5 won't be synchronised. Don't read what you didn't
> write!
> Logical volume "test_data" created.
>
> As far as output from lvm commands is concerned, it all looks fine to
> me. But why, when looking at lsblk is there so much strange alignment
> going on?Is this normal or should I be concerned? (I don't know what
> performance I should expect with my setup so I can't say anything
> about that). Note, that the same thing happens when I try to create a
> raid0
>
> # lsblk -t
> NAME ALIGNMENT MIN-IO OPT-IO PHY-SEC LOG-SEC ROTA
> SCHED RQ-SIZE RA WSAME
> sdc 0 4096 0 4096 512 1
> mq-deadline 60 128 32M
> ├─test-test_data_rmeta_0 0 4096 0 4096 512
> 1 128 128 32M
> │ └─test-test_data -1 65536 131072 4096 512
> 1 128 384 0B
> └─test-test_data_rimage_0 0 4096 0 4096 512
> 1 128 128 32M
> └─test-test_data -1 65536 131072 4096 512
> 1 128 384 0B
> sdd 0 4096 0 4096 512 1
> mq-deadline 60 128 32M
> ├─test-test_data_rmeta_1 0 4096 0 4096 512
> 1 128 128 32M
> │ └─test-test_data -1 65536 131072 4096 512
> 1 128 384 0B
> └─test-test_data_rimage_1 0 4096 0 4096 512
> 1 128 128 32M
> └─test-test_data -1 65536 131072 4096 512
> 1 128 384 0B
> sde 0 4096 0 4096 512 1
> mq-deadline 60 128 32M
> ├─test-test_data_rmeta_2 512 4096 0 4096 512
> 1 128 128 32M
> │ └─test-test_data -1 65536 131072 4096 512
> 1 128 384 0B
> └─test-test_data_rimage_2 512 4096 0 4096 512
> 1 128 128 32M
>
> # dmesg -wH
> [ +0.275210] device-mapper: raid: Superblocks created for new raid set
> [ +0.002796] md/raid:mdX: device dm-1 operational as raid disk 0
> [ +0.000004] md/raid:mdX: device dm-3 operational as raid disk 1
> [ +0.000002] md/raid:mdX: device dm-5 operational as raid disk 2
> [ +0.000801] md/raid:mdX: raid level 5 active with 3 out of 3
> devices, algorithm 2
> [ +0.010185] device-mapper: table: 254:6: adding target device dm-5
> caused an alignment inconsistency: physical_block_size=4096,
> logical_block_size=512, alignment_offset=512, start=0
> [ +0.000005] device-mapper: table: 254:6: adding target device dm-5
> caused an alignment inconsistency: physical_block_size=4096,
> logical_block_size=512, alignment_offset=512, start=0
> [ +0.000084] device-mapper: table: 254:6: adding target device dm-5
> caused an alignment inconsistency: physical_block_size=4096,
> logical_block_size=512, alignment_offset=512, start=0
> [ +0.000003] device-mapper: table: 254:6: adding target device dm-5
> caused an alignment inconsistency: physical_block_size=4096,
> logical_block_size=512, alignment_offset=512, start=0
> [ +0.070102] mdX: bitmap file is out of date, doing full recovery
>
>
> Platform:
> Arch Linux 5.6.0-rc7-1-git-00151-g67d584e33e54 (also tested with
> Debian on 4.19, same results)
> LVM version: 2.02.186(2) (2019-08-27)
> Library version: 1.02.164 (2019-08-27)
> Driver version: 4.42.0
>
>
> Any advice would be much appreciated, thanks
>
>
> _______________________________________________
> linux-lvm mailing list
> linux-lvm@redhat.com
> https://www.redhat.com/mailman/listinfo/linux-lvm
> read the LVM HOW-TO at http://tldp.org/HOWTO/LVM-HOWTO/
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2020-04-09 10:24 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-03-28 16:59 [linux-lvm] lvcreate: adding devices causes an alignment inconsistencies Bernhard Sulzer
2020-04-09 10:24 ` Heinz Mauelshagen
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).