{"id":2336,"date":"2016-07-11T12:57:38","date_gmt":"2016-07-11T00:57:38","guid":{"rendered":"http:\/\/www.zoyinc.com\/?p=2336"},"modified":"2016-07-13T06:08:16","modified_gmt":"2016-07-12T18:08:16","slug":"create-a-single-mirror-disk-for-raid1","status":"publish","type":"post","link":"http:\/\/www.zoyinc.com\/?p=2336","title":{"rendered":"Create a single mirror disk for a RAID1 array"},"content":{"rendered":"

I have a problem, I have a 3TB drive which has all my data on it but I need to move to a new RAID solution using Open Media Vault, OMV.<\/p>\n

I want to set up a mirrored, RAID1, 3TB array so will need two 3TB drives. I plan to use my current 3TB drive as half of the array. I could just buy two new 3TB disks but I can’t afford to do this.<\/p>\n

The plan is to create a mirrored array with just one disk, then copy my data from my old disk to my new one. Then wipe my old drive and add it to the array.<\/p>\n

This post is about how to create a mirrored RAID1 array in Open Media Vault but with only one disk. OMV doesn’t allow you to do this as it won’t allow you to create a RAID1 array with just one disk.<\/p>\n

Steps:<\/h2>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
1.<\/td>\nTo begin with run OMV with only the system disk attached.<\/td>\n<\/tr>\n
2.<\/td>\nSSH to your OMV server as root.<\/td>\n<\/tr>\n
3.<\/td>\nEdit the mdadm config file, \/etc\/mdadm\/mdadm.conf which should look like:<\/p>\n
# mdadm.conf\r\n#\r\n# Please refer to mdadm.conf(5) for information about this file.\r\n#\r\n\r\n# by default, scan all partitions (\/proc\/partitions) for MD superblocks.\r\n# alternatively, specify devices to scan, using wildcards if desired.\r\n# Note, if no DEVICE line is present, then \"DEVICE partitions\" is assumed.\r\n# To avoid the auto-assembly of RAID devices a pattern that CAN'T match is\r\n# used if no RAID devices are configured.\r\nDEVICE partitions\r\n\r\n# auto-create devices with Debian standard permissions\r\nCREATE owner=root group=disk mode=0660 auto=yes\r\n\r\n# automatically tag new arrays as belonging to the local system\r\nHOMEHOST <system>\r\n\r\n# definitions of existing MD arrays\r\nARRAY \/dev\/md0 metadata=1.2 name=nas2:RAID1Mirror UUID=a77b4c17:6ddd1afc:58bf7cdd:a51275ef\r\n\r\n# instruct the monitoring daemon where to send mail alerts\r\nMAILADDR tony@cantabrian.co.nz\r\nMAILFROM root\r\n<\/pre>\n
If you see any lines that look like:<\/p>\n
ARRAY \/dev\/md0 metadata=1.2 name=nas2:RAID1Mirror UUID=a77b4c17:6ddd1afc:58bf7cdd:a51275ef<\/pre>\n
Then you need to comment them out. The main issue seems to be if you create a new disk that clashes with an existing one around the device details, in this case “\/dev\/md0”. In my situation I am only having one RAID array on the box so it’s simple I don’t need any other array definitions. If you have multiple arrays you might want to play around a bit or do some Googling.<\/td>\n<\/tr>\n
4.<\/td>\nNow boot from a Debian Live USB disk, you can see how to create on in the post Create a Debian Live USB flash drive<\/a><\/td>\n<\/tr>\n
5.<\/td>\nIf you are considering installing mdadm, please don’t do it just yet. That will come later.<\/td>\n<\/tr>\n
6.<\/td>\nRun lsblk to work out which drive is the RAID one:<\/p>\n
root@debian:~# lsblk\r\nNAME\u00a0\u00a0 MAJ:MIN RM\u00a0 SIZE RO TYPE MOUNTPOINT\r\nsda\u00a0\u00a0\u00a0\u00a0\u00a0 8:0\u00a0\u00a0\u00a0 0\u00a0 2.7T\u00a0 0 disk\r\nsdb\u00a0\u00a0\u00a0\u00a0\u00a0 8:16\u00a0\u00a0 1\u00a0 7.5G\u00a0 0 disk\r\n\u2514\u2500sdb1\u00a0\u00a0 8:17\u00a0\u00a0 1\u00a0 1.2G\u00a0 0 part \/lib\/live\/mount\/medium\r\nloop0\u00a0\u00a0\u00a0 7:0\u00a0\u00a0\u00a0 0\u00a0 1.1G\u00a0 1 loop \/lib\/live\/mount\/rootfs\/filesystem.squashfs<\/pre>\n
From this you can see the new disk is sda – by looking at the size, 2.7 TB and also that sdb is mounted as “\/lib\/live\/mount\/medium”<\/td>\n<\/tr>\n
7.<\/td>\nNow we wipe the old disk just to make sure it is nice and clean, this is why it is good to remove the other disks:<\/p>\n

First run:<\/p>\n

sgdisk --zap \/dev\/sda<\/pre>\n
Then run “gdisk \/dev\/sda” and enter “x” at the first prompt followed by “z”. Finally when prompted “Blank out MBR” also enter “y”. This should all look like:<\/p>\n
root@debian:~# gdisk \/dev\/sda\r\nGPT fdisk (gdisk) version 0.8.10\r\n\r\nPartition table scan:\r\n\u00a0 MBR: not present\r\n\u00a0 BSD: not present\r\n\u00a0 APM: not present\r\n\u00a0 GPT: not present\r\n\r\nCreating new GPT entries.\r\n\r\nCommand (? for help): x<\/strong>\r\n\r\nExpert command (? for help): z<\/strong><\/span>\r\nAbout to wipe out GPT on \/dev\/sda. Proceed? (Y\/N): y\r\nGPT data structures destroyed! You may now partition the disk using fdisk or\r\nother utilities.\r\nBlank out MBR? (Y\/N): y<\/strong><\/span><\/pre>\n<\/td>\n<\/tr>\n
8.<\/td>\nNow run “fdisk -l \/dev\/sda” just to check everything went ok:<\/p>\n
root@debian:~# fdisk -l \/dev\/sda\r\n\r\nDisk \/dev\/sda: 2.7 TiB, 3000592982016 bytes, 5860533168 sectors\r\nUnits: sectors of 1 * 512 = 512 bytes\r\nSector size (logical\/physical): 512 bytes \/ 4096 bytes\r\nI\/O size (minimum\/optimal): 4096 bytes \/ 4096 bytes<\/pre>\n
So no partitions showing.<\/td>\n<\/tr>\n
9.<\/td>\nNow you can install mdadm<\/p>\n
apt-get install mdadm rsync initramfs-tools<\/pre>\n
During the install you will be asked the question \u201cMD arrays needed for the root file system\u201d. You need to enter \u201cnone\u201d !<\/p>\n
As has been mentioned before, ensure the window for the terminal is not too small otherwise when you are asked the question about “MD arrays needed for the root file system” you will only see:<\/p>\n
\"mdadm<\/p>\n
If you see the above, simply kill the terminal session and start again with a bigger window and you will now see:<\/p>\n
\"mdadm<\/td>\n<\/tr>\n
10.<\/td>\nNow create the array by running:<\/p>\n
mdadm --create \/dev\/md0 --level=1 --raid-devices=2 missing \/dev\/sda --name=nas2:Mirror --force<\/pre>\n
This will look like:<\/p>\n
root@debian:~# mdadm --create \/dev\/md0 --level=1 --raid-devices=2 missing \/dev\/sda --name=nas2:Mirror --force\r\nmdadm: Note: this array has metadata at the start and\r\n\u00a0\u00a0\u00a0 may not be suitable as a boot device.\u00a0 If you plan to\r\n\u00a0\u00a0\u00a0 store '\/boot' on this device please ensure that\r\n\u00a0\u00a0\u00a0 your boot-loader understands md\/v1.x metadata, or use\r\n\u00a0\u00a0\u00a0 --metadata=0.90\r\nContinue creating array? y\r\nmdadm: Defaulting to version 1.2 metadata\r\nmdadm: array \/dev\/md0 started.<\/pre>\n
Clearly the parameter “–name” is the name of the array. I would suggest you name it “<OMV hostname>:<raid name>” to fit in with what would be created if you had done it through OMV. So clearly in my case the hostname for the OMV server is “nas2”.<\/p>\n
Obviously you need to watch the “\/dev\/sda” matches your disk.<\/td>\n<\/tr>\n
11.<\/td>\nNow that the RAID array has been created you should add a file system so run “mkfs.ext4 -L Data \/dev\/md0”. The “-L” parameter is the name\/label for the file system, in my case “Data”. Also the “\/dev\/md0” must clearly match the details you entered when creating the array.<\/p>\n
root@debian:~# mkfs.ext4 -L Data \/dev\/md0\r\nmke2fs 1.42.12 (29-Aug-2014)\r\n\/dev\/md0 contains a ext4 file system labelled 'Data2'\r\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 last mounted on Mon Jul 11 08:46:39 2016\r\nProceed anyway? (y,n) y\r\nCreating filesystem with 732533872 4k blocks and 183140352 inodes\r\nFilesystem UUID: f20b47a4-d207-40aa-a7e6-42ea82354455\r\nSuperblock backups stored on blocks:\r\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208,\r\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 4096000, 7962624, 11239424, 20480000, 23887872, 71663616, 78675968,\r\n\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 102400000, 214990848, 512000000, 550731776, 644972544\r\n\r\nAllocating group tables: done\r\nWriting inode tables: done\r\nCreating journal (32768 blocks): done\r\nWriting superblocks and filesystem accounting information: done<\/pre>\n<\/td>\n<\/tr>\n
12.<\/td>\nShutdown the Deblian live and connect your disks back and boot into OMV<\/td>\n<\/tr>\n
13.<\/td>\nNow when you boot into OMV if you select “Storage | File Systems” you will see your new file system, we called “Data”, but it won’t be mounted:<\/p>\n

\"OMV<\/p>\n

You should select the “Data” file system and click on mount. After you have done that you will get the “The configuration has changed message…” and you should click on “Apply”.<\/td>\n<\/tr>\n

14.<\/td>\nFrom your terminal session you should add the array to the mdadm.conf:<\/p>\n
mdadm --detail --scan >> \/etc\/mdadm\/mdadm.conf<\/pre>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
 <\/p>\n
You should now be in a position to create shares and so on.<\/p>\n
 <\/p>\n
Resources<\/h2>\n
Setting up RAID on an existing Debian\/Ubuntu installation<\/span><\/p>\n
https:\/\/feeding.cloud.geek.nz\/posts\/setting-up-raid-on-existing\/<\/a><\/p>\n
 <\/p>\n
Mdadm Cheat Sheet<\/p>\n
http:\/\/www.ducea.com\/2009\/03\/08\/mdadm-cheat-sheet\/<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"
I have a problem, I have a 3TB drive which has all my data on it but I need to move to a new RAID solution using Open Media Vault, OMV. I want to set up a mirrored, RAID1, 3TB array so will need two 3TB drives. I plan to use my current 3TB drive […]<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[312,313],"tags":[318,338,337,319,320,326],"class_list":["post-2336","post","type-post","status-publish","format-standard","hentry","category-debian","category-open-media-vault","tag-debian","tag-mdadm","tag-missing","tag-omv","tag-open-media-vault","tag-raid"],"_links":{"self":[{"href":"http:\/\/www.zoyinc.com\/index.php?rest_route=\/wp\/v2\/posts\/2336","targetHints":{"allow":["GET"]}}],"collection":[{"href":"http:\/\/www.zoyinc.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/www.zoyinc.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/www.zoyinc.com\/index.php?rest_route=\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"http:\/\/www.zoyinc.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=2336"}],"version-history":[{"count":23,"href":"http:\/\/www.zoyinc.com\/index.php?rest_route=\/wp\/v2\/posts\/2336\/revisions"}],"predecessor-version":[{"id":2506,"href":"http:\/\/www.zoyinc.com\/index.php?rest_route=\/wp\/v2\/posts\/2336\/revisions\/2506"}],"wp:attachment":[{"href":"http:\/\/www.zoyinc.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=2336"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/www.zoyinc.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=2336"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/www.zoyinc.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=2336"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}