Roman Dissauer
2015-03-19 07:00:06 UTC
Hi all,
I’m on centos 7 and have several logical volumes on my machine. Yesterday I scrubbed the raid1/raid10 logical volumes for the first time after 6 months. Scrubbing reported multi million mismatches on three of my logical volumes (the older ones). Recently created lvs didn’t have any mismatches though. Now I raised some questions:
1. Is this a normal behaviour of raid1/raid10 to have this much mismatches?
2. Is it safe to do "lvchange --syncaction repair“ on this logical volumes?
3. How does lvm know which one is the „good“ side of the raid1?
4. Is it a good idea to do „lvchange --resync“ from time to time?
5. Is it more reliable to change the lvm from raid10 to raid6?
maybe someone can help me answering my questions.
enclosed you’ll find some more information.
Thanks,
Roman Dissauer
Here is the scrub output:
LV VG Attr Start SSize #Str Type Stripe Chunk Mismatches SyncAction Cpy%Sync
data vg0 rwi-aor-m- 0 3,49t 4 raid10 512,00k 0 895444480 idle 100,00
owncloud vg0 rwi-aor-m- 0 250,00g 2 raid1 0 0 285961472 idle 100,00
root vg0 rwi-aor-m- 0 30,00g 4 raid1 0 0 16501376 idle 100,00
on data this is about 0.1% mismatch of the volume
on owncloud this is about 0.4% mismatch of the volume
on root this is about 0.2% mismatch of the volume
And here the output of the system storage manager:
---------------------------------------------------------------
Device Free Used Total Pool Mount point
---------------------------------------------------------------
/dev/sda 2.73 TB PARTITIONED
/dev/sda1 873.50 GB 1.88 TB 2.73 TB vg0
/dev/sdb 2.73 TB
/dev/sdb1 860.84 GB 1.89 TB 2.73 TB vg0
/dev/sdc 2.73 TB
/dev/sdc1 10.82 GB 2.72 TB 2.73 TB vg0
/dev/sdd 2.73 TB
/dev/sdd1 10.82 GB 2.72 TB 2.73 TB vg0
/dev/sde 7.32 GB
/dev/sde1 500.00 MB /boot
/dev/sdg 0.00 KB 931.51 GB 931.51 GB backup
/dev/sdh 0.00 KB 2.73 TB 2.73 TB backup
---------------------------------------------------------------
-------------------------------------------------
Pool Type Devices Free Used Total
-------------------------------------------------
backup lvm 2 0.00 KB 3.64 TB 3.64 TB
vg0 lvm 4 1.71 TB 9.20 TB 10.92 TB
-------------------------------------------------
-------------------------------------------------------------------------------------------
Volume Pool Volume size FS FS size Free Type Mount point
-------------------------------------------------------------------------------------------
/dev/backup/backup backup 3.64 TB ext4 3.64 TB 509.38 GB linear /backup
/dev/vg0/data vg0 3.49 TB ext4 3.49 TB 595.34 GB raid10 /data
/dev/vg0/home vg0 200.00 GB ext4 200.00 GB 128.58 GB raid10 /home
/dev/vg0/owncloud vg0 250.00 GB xfs 249.99 GB 145.01 GB raid1 /var/owncloud
/dev/vg0/jail vg0 10.00 GB ext4 10.00 GB 6.71 GB raid1 /jail
/dev/vg0/fotos vg0 500.00 GB ext4 500.00 GB 456.07 GB raid1 /fotos
/dev/vg0/virtswap vg0 20.00 GB ext4 20.00 GB 17.89 GB striped /virt/swap
/dev/vg0/iso vg0 10.00 GB ext4 10.00 GB 8.05 GB linear /virt/iso
/dev/vg0/images vg0 100.00 GB ext4 100.00 GB 83.09 GB raid1 /virt/images
/dev/vg0/root vg0 30.00 GB xfs 29.99 GB 5.79 GB raid1 /
/dev/vg0/swap vg0 8.00 GB striped
/dev/sde1 500.00 MB xfs 496.67 MB 240.59 MB /boot
-------------------------------------------------------------------------------------------
Disks are WD RED 3TB
and here some revision numbers:
uname -r
3.10.0-123.20.1.el7.x86_64
lvm version
LVM version: 2.02.105(2)-RHEL7 (2014-03-26)
Library version: 1.02.84-RHEL7 (2014-03-26)
Driver version: 4.27.0
I’m on centos 7 and have several logical volumes on my machine. Yesterday I scrubbed the raid1/raid10 logical volumes for the first time after 6 months. Scrubbing reported multi million mismatches on three of my logical volumes (the older ones). Recently created lvs didn’t have any mismatches though. Now I raised some questions:
1. Is this a normal behaviour of raid1/raid10 to have this much mismatches?
2. Is it safe to do "lvchange --syncaction repair“ on this logical volumes?
3. How does lvm know which one is the „good“ side of the raid1?
4. Is it a good idea to do „lvchange --resync“ from time to time?
5. Is it more reliable to change the lvm from raid10 to raid6?
maybe someone can help me answering my questions.
enclosed you’ll find some more information.
Thanks,
Roman Dissauer
Here is the scrub output:
LV VG Attr Start SSize #Str Type Stripe Chunk Mismatches SyncAction Cpy%Sync
data vg0 rwi-aor-m- 0 3,49t 4 raid10 512,00k 0 895444480 idle 100,00
owncloud vg0 rwi-aor-m- 0 250,00g 2 raid1 0 0 285961472 idle 100,00
root vg0 rwi-aor-m- 0 30,00g 4 raid1 0 0 16501376 idle 100,00
on data this is about 0.1% mismatch of the volume
on owncloud this is about 0.4% mismatch of the volume
on root this is about 0.2% mismatch of the volume
And here the output of the system storage manager:
---------------------------------------------------------------
Device Free Used Total Pool Mount point
---------------------------------------------------------------
/dev/sda 2.73 TB PARTITIONED
/dev/sda1 873.50 GB 1.88 TB 2.73 TB vg0
/dev/sdb 2.73 TB
/dev/sdb1 860.84 GB 1.89 TB 2.73 TB vg0
/dev/sdc 2.73 TB
/dev/sdc1 10.82 GB 2.72 TB 2.73 TB vg0
/dev/sdd 2.73 TB
/dev/sdd1 10.82 GB 2.72 TB 2.73 TB vg0
/dev/sde 7.32 GB
/dev/sde1 500.00 MB /boot
/dev/sdg 0.00 KB 931.51 GB 931.51 GB backup
/dev/sdh 0.00 KB 2.73 TB 2.73 TB backup
---------------------------------------------------------------
-------------------------------------------------
Pool Type Devices Free Used Total
-------------------------------------------------
backup lvm 2 0.00 KB 3.64 TB 3.64 TB
vg0 lvm 4 1.71 TB 9.20 TB 10.92 TB
-------------------------------------------------
-------------------------------------------------------------------------------------------
Volume Pool Volume size FS FS size Free Type Mount point
-------------------------------------------------------------------------------------------
/dev/backup/backup backup 3.64 TB ext4 3.64 TB 509.38 GB linear /backup
/dev/vg0/data vg0 3.49 TB ext4 3.49 TB 595.34 GB raid10 /data
/dev/vg0/home vg0 200.00 GB ext4 200.00 GB 128.58 GB raid10 /home
/dev/vg0/owncloud vg0 250.00 GB xfs 249.99 GB 145.01 GB raid1 /var/owncloud
/dev/vg0/jail vg0 10.00 GB ext4 10.00 GB 6.71 GB raid1 /jail
/dev/vg0/fotos vg0 500.00 GB ext4 500.00 GB 456.07 GB raid1 /fotos
/dev/vg0/virtswap vg0 20.00 GB ext4 20.00 GB 17.89 GB striped /virt/swap
/dev/vg0/iso vg0 10.00 GB ext4 10.00 GB 8.05 GB linear /virt/iso
/dev/vg0/images vg0 100.00 GB ext4 100.00 GB 83.09 GB raid1 /virt/images
/dev/vg0/root vg0 30.00 GB xfs 29.99 GB 5.79 GB raid1 /
/dev/vg0/swap vg0 8.00 GB striped
/dev/sde1 500.00 MB xfs 496.67 MB 240.59 MB /boot
-------------------------------------------------------------------------------------------
Disks are WD RED 3TB
and here some revision numbers:
uname -r
3.10.0-123.20.1.el7.x86_64
lvm version
LVM version: 2.02.105(2)-RHEL7 (2014-03-26)
Library version: 1.02.84-RHEL7 (2014-03-26)
Driver version: 4.27.0