我使用 4 个 75G 驱动器创建了一个 RAID10 阵列，以创建 150G 的存储空间。
一切完成后（包括初始同步），一切看起来都很好，只是输出df -h显示指定挂载点上只有 73G 存储空间。
细节：
- 该机器是 Amazon EC2 上的 m1.large Ubuntu 11.10 实例。
- 这 4 个驱动器是 EBS 驱动器，每个驱动器大小为 75G。
- RAID10 阵列是使用以下脚本创建的：
-
#!/bin/sh
disk1="/dev/sdh1"
disk2="/dev/sdh2"
disk3="/dev/sdh3"
disk4="/dev/sdh4"
echo "*** Verifying existence of 4 volumes $disk1, $disk2, $disk3 and $disk4"
if [ -b "$disk1" -a -b "$disk2" -a -b "$disk3" -a -b "$disk4" ]; then
echo "# Found expected block devices."
else
echo "!!! Did not find expected block devices. Error."
exit -1
fi
until read -p "??? - How big (in GB) are the disks (They should be the same size)? " disk_size && [ $disk_size ]; do
echo "Please enter a disk size."
done
lv_size=$(echo "scale=2; $disk_size * 2.0" | bc)
echo "*** Assuming a per disk size of $disk_size gigs, will create a logical volume of $lv_size gigs, with $lv_size reserved for snapshots"
echo "*** Partitioning disks..."
echo "~ Partitioning $disk1"
echo ',,L' | sfdisk $disk1
echo "~ Partitioning $disk2"
echo ',,L' | sfdisk $disk2
echo "~ Partitioning $disk3"
echo ',,L' | sfdisk $disk3
echo "~ Partitioning $disk4"
echo ',,L' | sfdisk $disk4
sleep 6
echo "*** Creating /dev/md0 as a RAID 10"
/sbin/mdadm /dev/md0 --create --level=10 --raid-devices=4 $disk1 $disk2 $disk3 $disk4
echo " ~ Allocating /dev/md0 as a physical volume."
/sbin/pvcreate /dev/md0
echo " ~ Allocating a Volume Group 'mongodb_vg'"
/sbin/vgcreate -s 64M mongodb_vg /dev/md0
echo " ~ Creating a Logical Volume 'mongodb_lv'"
num_extents=$(echo "$disk_size * 1000 / 64" | bc)
/sbin/lvcreate -l $num_extents -nmongodb_lv mongodb_vg
echo " ~ Formatting the new volume (/dev/mongodb_vg/mongodb_lv) with EXT4"
/sbin/mkfs.ext4 /dev/mongodb_vg/mongodb_lv
echo " ~ Done! Go ahead and mount the new filesystem. Suggested FStab: "
echo " /dev/mongodb_vg/mongodb_lv /data ext4 defaults,noatime 0 0"
这是我得到的输出：
*** Verifying existence of 4 volumes /dev/xvdh1, /dev/xvdh2, /dev/xvdh3 and /dev/xvdh4
# Found expected block devices.
??? - How big (in GB) are the disks (They should be the same size)? 75
*** Assuming a per disk size of 75 gigs, will create a logical volume of 150.0 gigs, with 150.0 reserved for snapshots
*** Partitioning disks...
~ Partitioning /dev/xvdh1
Checking that no-one is using this disk right now ...
BLKRRPART: Invalid argument
OK
Disk /dev/xvdh1: 9790 cylinders, 255 heads, 63 sectors/track
sfdisk: ERROR: sector 0 does not have an msdos signature
/dev/xvdh1: unrecognized partition table type
Old situation:
No partitions found
New situation:
Units = cylinders of 8225280 bytes, blocks of 1024 bytes, counting from 0
Device Boot Start End #cyls #blocks Id System
/dev/xvdh1p1 0+ 9789 9790- 78638174+ 83 Linux
/dev/xvdh1p2 0 - 0 0 0 Empty
/dev/xvdh1p3 0 - 0 0 0 Empty
/dev/xvdh1p4 0 - 0 0 0 Empty
Warning: no primary partition is marked bootable (active)
This does not matter for LILO, but the DOS MBR will not boot this disk.
Successfully wrote the new partition table
Re-reading the partition table ...
BLKRRPART: Invalid argument
If you created or changed a DOS partition, /dev/foo7, say, then use dd(1)
to zero the first 512 bytes: dd if=/dev/zero of=/dev/foo7 bs=512 count=1
(See fdisk(8).)
~ Partitioning /dev/xvdh2
Checking that no-one is using this disk right now ...
BLKRRPART: Invalid argument
OK
Disk /dev/xvdh2: 9790 cylinders, 255 heads, 63 sectors/track
sfdisk: ERROR: sector 0 does not have an msdos signature
/dev/xvdh2: unrecognized partition table type
Old situation:
No partitions found
New situation:
Units = cylinders of 8225280 bytes, blocks of 1024 bytes, counting from 0
Device Boot Start End #cyls #blocks Id System
/dev/xvdh2p1 0+ 9789 9790- 78638174+ 83 Linux
/dev/xvdh2p2 0 - 0 0 0 Empty
/dev/xvdh2p3 0 - 0 0 0 Empty
/dev/xvdh2p4 0 - 0 0 0 Empty
Warning: no primary partition is marked bootable (active)
This does not matter for LILO, but the DOS MBR will not boot this disk.
Successfully wrote the new partition table
Re-reading the partition table ...
BLKRRPART: Invalid argument
If you created or changed a DOS partition, /dev/foo7, say, then use dd(1)
to zero the first 512 bytes: dd if=/dev/zero of=/dev/foo7 bs=512 count=1
(See fdisk(8).)
~ Partitioning /dev/xvdh3
Checking that no-one is using this disk right now ...
BLKRRPART: Invalid argument
OK
Disk /dev/xvdh3: 9790 cylinders, 255 heads, 63 sectors/track
sfdisk: ERROR: sector 0 does not have an msdos signature
/dev/xvdh3: unrecognized partition table type
Old situation:
No partitions found
New situation:
Units = cylinders of 8225280 bytes, blocks of 1024 bytes, counting from 0
Device Boot Start End #cyls #blocks Id System
/dev/xvdh3p1 0+ 9789 9790- 78638174+ 83 Linux
/dev/xvdh3p2 0 - 0 0 0 Empty
/dev/xvdh3p3 0 - 0 0 0 Empty
/dev/xvdh3p4 0 - 0 0 0 Empty
Warning: no primary partition is marked bootable (active)
This does not matter for LILO, but the DOS MBR will not boot this disk.
Successfully wrote the new partition table
Re-reading the partition table ...
BLKRRPART: Invalid argument
If you created or changed a DOS partition, /dev/foo7, say, then use dd(1)
to zero the first 512 bytes: dd if=/dev/zero of=/dev/foo7 bs=512 count=1
(See fdisk(8).)
~ Partitioning /dev/xvdh4
Checking that no-one is using this disk right now ...
BLKRRPART: Invalid argument
OK
Disk /dev/xvdh4: 9790 cylinders, 255 heads, 63 sectors/track
sfdisk: ERROR: sector 0 does not have an msdos signature
/dev/xvdh4: unrecognized partition table type
Old situation:
No partitions found
New situation:
Units = cylinders of 8225280 bytes, blocks of 1024 bytes, counting from 0
Device Boot Start End #cyls #blocks Id System
/dev/xvdh4p1 0+ 9789 9790- 78638174+ 83 Linux
/dev/xvdh4p2 0 - 0 0 0 Empty
/dev/xvdh4p3 0 - 0 0 0 Empty
/dev/xvdh4p4 0 - 0 0 0 Empty
Warning: no primary partition is marked bootable (active)
This does not matter for LILO, but the DOS MBR will not boot this disk.
Successfully wrote the new partition table
Re-reading the partition table ...
BLKRRPART: Invalid argument
If you created or changed a DOS partition, /dev/foo7, say, then use dd(1)
to zero the first 512 bytes: dd if=/dev/zero of=/dev/foo7 bs=512 count=1
(See fdisk(8).)
*** Creating /dev/md0 as a RAID 10
mdadm: partition table exists on /dev/xvdh1 but will be lost or
meaningless after creating array
mdadm: partition table exists on /dev/xvdh2 but will be lost or
meaningless after creating array
mdadm: partition table exists on /dev/xvdh3 but will be lost or
meaningless after creating array
mdadm: partition table exists on /dev/xvdh4 but will be lost or
meaningless after creating array
Continue creating array? y
mdadm: Defaulting to version 1.2 metadata
mdadm: array /dev/md0 started.
~ Allocating /dev/md0 as a physical volume.
Physical volume "/dev/md0" successfully created
~ Allocating a Volume Group 'mongodb_vg'
Volume group "mongodb_vg" successfully created
~ Creating a Logical Volume 'mongodb_lv'
Logical volume "mongodb_lv" created
~ Formatting the new volume (/dev/mongodb_vg/mongodb_lv) with EXT4
mke2fs 1.41.14 (22-Dec-2010)
Filesystem label=
OS type: Linux
Block size=4096 (log=2)
Fragment size=4096 (log=2)
Stride=128 blocks, Stripe width=256 blocks
4800512 inodes, 19185664 blocks
959283 blocks (5.00%) reserved for the super user
First data block=0
Maximum filesystem blocks=4294967296
586 block groups
32768 blocks per group, 32768 fragments per group
8192 inodes per group
Superblock backups stored on blocks:
32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208,
4096000, 7962624, 11239424
Writing inode tables: done
Creating journal (32768 blocks): done
Writing superblocks and filesystem accounting information: done
This filesystem will be automatically checked every 36 mounts or
180 days, whichever comes first. Use tune2fs -c or -i to override.
~ Done! Go ahead and mount the new filesystem. Suggested FStab:
/dev/mongodb_vg/mongodb_lv /data ext4 defaults,noatime 0 0
这是相关输出df -h：
Filesystem Size Used Avail Use% Mounted on
/dev/mapper/mongodb_vg-mongodb_lv
73G 180M 69G 1% /ebsRaid
这是输出mdadm --detail /dev/md0
/dev/md0:
Version : 1.2
Creation Time : Wed Feb 29 10:14:39 2012
Raid Level : raid10
Array Size : 157283328 (150.00 GiB 161.06 GB)
Used Dev Size : 78641664 (75.00 GiB 80.53 GB)
Raid Devices : 4
Total Devices : 4
Persistence : Superblock is persistent
Update Time : Wed Feb 29 13:21:49 2012
State : clean
Active Devices : 4
Working Devices : 4
Failed Devices : 0
Spare Devices : 0
Layout : near=2
Chunk Size : 512K
Name : my.site.com:0 (local to host my.site.com)
UUID : CENSORED
Events : 19
Number Major Minor RaidDevice State
0 202 113 0 active sync /dev/xvdh1
1 202 114 1 active sync /dev/xvdh2
2 202 115 2 active sync /dev/xvdh3
3 202 116 3 active sync /dev/xvdh4
这是输出cat /proc/mdstat：
Personalities : [raid10]
md0 : active raid10 xvdh4[3] xvdh3[2] xvdh2[1] xvdh1[0]
157283328 blocks super 1.2 512K chunks 2 near-copies [4/4] [UUUU]
unused devices: <none>
编辑1：
这是 lvdisplay -m 的输出：
--- Logical volume ---
LV Name /dev/mongodb_vg/mongodb_lv
VG Name mongodb_vg
LV UUID SEpGth-cXd3-ZFhy-XLHo-T5pV-gEd1-Tgancs
LV Write Access read/write
LV Status available
# open 1
LV Size 73.19 GiB
Current LE 1171
Segments 1
Allocation inherit
Read ahead sectors auto
- currently set to 4096
Block device 252:0
--- Segments ---
Logical extent 0 to 1170:
Type linear
Physical volume /dev/md0
Physical extents 0 to 1170
编辑2：
这是输出vgdisplay：
--- Volume group ---
VG Name mongodb_vg
System ID
Format lvm2
Metadata Areas 1
Metadata Sequence No 2
VG Access read/write
VG Status resizable
MAX LV 0
Cur LV 1
Open LV 1
Max PV 0
Cur PV 1
Act PV 1
VG Size 149.94 GiB
PE Size 64.00 MiB
Total PE 2399
Alloc PE / Size 1171 / 73.19 GiB
Free PE / Size 1228 / 76.75 GiB
VG UUID CENSORED
答案1
您的卷组未使用为其创建的全部范围：
VG Size 149.94 GiB
PE Size 64.00 MiB
Total PE 2399
Alloc PE / Size 1171 / 73.19 GiB
Free PE / Size 1228 / 76.75 GiB
您可以使用以下命令添加更多范围：
lvextend -l +100%FREE /dev/mongodb_vg/mongodb_lv /dev/md0
输入之前请先阅读手册页
此命令将扩展卷组以使用所有剩余的可用扩展区（如果您想保留一些可用扩展区，也可以选择较少的扩展区）。它将利用 md0 上的扩展区来执行此操作。
然后您可以使用以下方法在线调整分区大小：
resize2fs /dev/mongodb_vg/mongodb_lv
并且它应该显示正在在线调整大小。我相信这会解决您的问题，但在尝试之前请阅读手册页并了解它们的作用。我不会对您损坏磁盘负责。
顺便说一句，在 EBS 上安装 RAID，然后在其上安装 lvm，这似乎是不必要的磁盘虚拟化级别。添加额外的 RAID 不会提高性能，也不会提高数据的安全性。如果我没记错的话，LVM 已经可以进行镜像/条带化了。虽然从技术上讲，您可以在 RAID 上的 LVM 上无限次地运行 RAID 上的 LVM，但我不确定这样做是否能获得很多好处（尽管我很高兴有人指出我在这方面是错的）。