LVM : 論理ボリュームのミラーリング2021/03/08 |
|
ミラーリング論理ボリュームの作成方法です。
|
|
| [1] | 例として、物理ボリューム [/dev/sdb1] と [/dev/sdc1] を使用してミラーリングボリュームを作成します。 |
|
# [/dev/sdb1] と [/dev/sdc1] でボリュームグループ [vg_mirror] を作成 [root@dlp ~]# vgcreate vg_mirror /dev/sdb1 /dev/sdc1 Volume group "vg_mirror" successfully created # ミラー論理ボリューム作成 [root@dlp ~]# lvcreate -L 50G -m1 -n lv_mirror vg_mirror [ 1101.040596] raid6: sse2x1 gen() 11960 MB/s [ 1101.058579] raid6: sse2x1 xor() 7586 MB/s [ 1101.076584] raid6: sse2x2 gen() 13040 MB/s [ 1101.094582] raid6: sse2x2 xor() 9199 MB/s [ 1101.112586] raid6: sse2x4 gen() 13943 MB/s [ 1101.130587] raid6: sse2x4 xor() 5827 MB/s [ 1101.148585] raid6: avx2x1 gen() 14740 MB/s [ 1101.166583] raid6: avx2x1 xor() 13388 MB/s [ 1101.184582] raid6: avx2x2 gen() 18628 MB/s [ 1101.202580] raid6: avx2x2 xor() 16235 MB/s [ 1101.220579] raid6: avx2x4 gen() 20438 MB/s [ 1101.238564] raid6: avx2x4 xor() 5846 MB/s [ 1101.239618] raid6: using algorithm avx2x4 gen() 20438 MB/s [ 1101.240966] raid6: .... xor() 5846 MB/s, rmw enabled [ 1101.242189] raid6: using avx2x2 recovery algorithm [ 1101.244508] async_tx: api initialized (async) [ 1101.246846] xor: automatically using best checksumming function avx [ 1101.267096] device-mapper: raid: Loading target version 1.15.1 [ 1101.319169] device-mapper: raid: Superblocks created for new raid set [ 1101.330455] md/raid1:mdX: not clean -- starting background reconstruction [ 1101.331515] md/raid1:mdX: active with 2 out of 2 mirrors [ 1101.463414] mdX: bitmap file is out of date, doing full recovery [ 1101.483596] md: resync of RAID array mdX Logical volume "lv_mirror" created.[root@dlp ~]# lvdisplay /dev/vg_mirror/lv_mirror --- Logical volume --- LV Path /dev/vg_mirror/lv_mirror LV Name lv_mirror VG Name vg_mirror LV UUID Cg00Sz-UwGv-Te4S-Xn1p-MJcD-AXwd-Xt1mDU LV Write Access read/write LV Creation host, time dlp.srv.world, 2021-03-07 16:22:32 +0900 LV Status available # open 0 LV Size 50.00 GiB Current LE 12800 Mirrored volumes 2 Segments 1 Allocation inherit Read ahead sectors auto - currently set to 8192 Block device 253:6 |
| [2] | すでに稼働中の論理ボリュームをミラー化する場合は以下のように設定します。 |
|
# ボリュームグループ拡張 [root@dlp ~]# vgextend vg_data /dev/sdc1 Volume group "vg_data" successfully extended # 論理ボリュームミラー化 [root@dlp ~]# lvconvert -m1 /dev/vg_data/lv_data /dev/sdc1
Are you sure you want to convert linear LV vg_data/lv_data to raid1 with 2 images enhancing resilience? [y/n]: y
[ 1386.352582] device-mapper: raid: Device 1 specified for rebuild; clearing superblock
[ 1386.354375] device-mapper: raid: Superblocks created for new raid set
[ 1386.363010] md/raid1:mdX: active with 1 out of 2 mirrors
[ 1386.496302] mdX: bitmap file is out of date, doing full recovery
[ 1386.517129] md: recovery of RAID array mdX
Logical volume vg_data/lv_data successfully converted.
# 確認 : [Cpy%Sync] が [100] でデータ同期完了 [root@dlp ~]# lvs LV VG Attr LSize Pool Origin Data% Meta% Move Log Cpy%Sync Convert root cs -wi-ao---- 71.00g swap cs -wi-ao---- 7.90g lv_data vg_data rwi-a-r--- 50.00g 13.12[root@dlp ~]# lvs LV VG Attr LSize Pool Origin Data% Meta% Move Log Cpy%Sync Convert root cs -wi-ao---- 71.00g swap cs -wi-ao---- 7.90g lv_data vg_data rwi-a-r--- 50.00g 100.00 |
| [3] | 論理ボリュームのミラー化を解除する場合は以下のように設定します。 |
|
# -m0 で解除 [root@dlp ~]# lvconvert -m0 /dev/vg_data/lv_data
Are you sure you want to convert raid1 LV vg_data/lv_data to type linear losing all resilience? [y/n]: y
Logical volume vg_data/lv_data successfully converted.
[root@dlp ~]# lvs -a -o vg_name,name,devices,size VG LV Devices LSize cs root /dev/sda2(7994) 71.00g cs root /dev/sda2(2023) 71.00g cs swap /dev/sda2(0) 7.90g vg_data lv_data /dev/sdb1(0) 50.00g |
| Sponsored Link |