我正在使用虚拟机来模拟。
md1 很慢，而 md2 很快。
我知道我正在跨 VG 执行此操作，并且如果 LV_Data 和 Lv_Cache 在同一个 VG 中，它就可以工作，但这没有任何意义，缓存在同一个磁盘（raid）中？不。
我发现了这个： https://bugzilla.redhat.com/show_bug.cgi?id=1503693#c0 我不确定是不是同样的原因
问题是：会发生什么？或者有其他方法可以解决这个问题？
详细信息如下：
[root@localhost ~]# pvdisplay
--- Physical volume ---
PV Name /dev/md1
VG Name vg_1
PV Size <12.00 TiB / not usable 5.00 MiB
Allocatable yes
PE Size 4.00 MiB
Total PE 3145531
Free PE 3145531
Allocated PE 0
PV UUID G1KYfG-oCfA-B2Oz-ckJH-Ozxa-Hvhq-gNzQKV
--- Physical volume ---
PV Name /dev/md2
VG Name vg_2
PV Size 255.87 GiB / not usable <2.94 MiB
Allocatable yes
PE Size 4.00 MiB
Total PE 65503
Free PE 65503
Allocated PE 0
PV UUID CmBVAe-EgPh-IzGC-8ew0-x5al-4r0H-htd2Ex
[root@localhost ~]# vgdisplay
--- Volume group ---
VG Name vg_1
System ID
Format lvm2
Metadata Areas 1
Metadata Sequence No 12
VG Access read/write
VG Status resizable
MAX LV 0
Cur LV 0
Open LV 0
Max PV 0
Cur PV 1
Act PV 1
VG Size <12.00 TiB
PE Size 4.00 MiB
Total PE 3145531
Alloc PE / Size 0 / 0
Free PE / Size 3145531 / <12.00 TiB
VG UUID I5eDgM-xNtS-eaCE-81l5-4tRL-YSb9-yNkbXm
--- Volume group ---
VG Name vg_2
System ID
Format lvm2
Metadata Areas 1
Metadata Sequence No 17
VG Access read/write
VG Status resizable
MAX LV 0
Cur LV 0
Open LV 0
Max PV 0
Cur PV 1
Act PV 1
VG Size 255.87 GiB
PE Size 4.00 MiB
Total PE 65503
Alloc PE / Size 0 / 0
Free PE / Size 65503 / 255.87 GiB
VG UUID l9vIDi-5yVp-0QCx-Rc9K-w7c4-WYOP-IfemFY
[root@localhost ~]# lvcreate -L 1T -n slow /dev/vg_1
Logical volume "slow" created.
[root@localhost ~]# lvcreate -L 250G -n fast /dev/vg_2
Logical volume "fast" created.
[root@localhost ~]# lvcreate -L 250M -n meta /dev/vg_2
Rounding up size to full physical extent 252.00 MiB
Logical volume "meta" created.
[root@localhost ~]# lvs
LV VG Attr LSize Pool Origin Data% Meta% Move Log Cpy%Sync Convert
slow vg_1 -wi-a----- 1.00t
fast vg_2 -wi-a----- 250.00g
meta vg_2 -wi-a----- 252.00m
[root@localhost ~]# lvconvert --type cache-pool --poolmetadata vg_2/meta vg_2/fast
Using 288.00 KiB chunk size instead of default 64.00 KiB, so cache pool has less then 1000000 chunks.
WARNING: Converting logical volume vg_2/fast and vg_2/meta to cache pool's data and metadata volumes with metadata wiping.
THIS WILL DESTROY CONTENT OF LOGICAL VOLUME (filesystem etc.)
Do you really want to convert vg_2/fast and vg_2/meta? [y/n]: y
Converted vg_2/fast_cdata to cache pool.
[root@localhost ~]# lvs -a
LV VG Attr LSize Pool Origin Data% Meta% Move Log Cpy%Sync Convert
slow vg_1 -wi-a----- 1.00t
fast vg_2 Cwi---C--- 250.00g
[fast_cdata] vg_2 Cwi------- 250.00g
[fast_cmeta] vg_2 ewi------- 252.00m
[lvol0_pmspare] vg_2 ewi------- 252.00m <=========== ? ( this didn't appear in the guide )
[root@localhost ~]# lvconvert --type cache --cachepool vg_2/fast vg_1/slow
VG name mismatch from position arg (vg_1) and option arg (vg_2). <=========== ? ( the error )
答案1
在同一 VG 中并不意味着在同一磁盘上。只能组合卷组内的 LV。
任何 lvcreate 都允许指定用于分配空间的物理卷。
所以你要：
在较大但较慢的 PV (/dev/md1) 上创建数据 LV
lvcreate -n data -L 2T vg /dev/md1在较小且较快的 PV（/dev/md2）上创建缓存池
lvcreate --type cache-pool -n cache -L 80G vg /dev/md2将数据 LV 转换为缓存
lvconvert --type cache --cachepool cache vg/data
或者分两步：跳过步骤 1，并使用以下方法代替步骤 3：
lvcreate --type cache -L 2T --cachepool cache vg /dev/md1
更多内容请参见man lvmcache。