In these days we can find lot of file systems and they are also used in many Operating systems and block devices . Comparing with this file systems BTRFS is a newbie . Btrfs is a new copy on write (CoW) filesystem for Linux aimed at implementing advanced features while focusing on fault tolerance, repair and easy administration.<\/p>\n
Recently CentOS\/RHEL is also supports BTRFS .<\/p>\n
So lets start the tutorial ,<\/p>\n
Consider we have one disk \/dev\/sdb .<\/p>\n
# fdisk \/dev\/sdb<\/em><\/p><\/blockquote>\n
And create disk called \/dev\/sdb1 .<\/em><\/p>\n
Format it using BTRFS .<\/p>\n
# mkfs.btrfs<\/span> \/dev\/sdb1<\/em><\/p><\/blockquote>\n
For testing purpose we are mounting this to \/mnt .<\/strong><\/p>\n
# mount \/dev\/sdb1 \/mnt<\/em><\/p><\/blockquote>\n
Now , BTRFS is mounted under \/mnt . Just check with df<\/em> <\/strong>command .<\/p>\n
Basically BTRFS is working with subvolume\u2019s . First we need to create a subvolume . Fo that we don\u2019t need to unmount file system .<\/p>\n
# btrfs sub create \/mnt\/volume1<\/em><\/p>\n
# tree \/mnt<\/em><\/p>\n
# btrfs sub list \/mnt<\/em><\/p><\/blockquote>\n
Now we need to remount with \u201cvolume1\u201d as our default mount mount. For that ,<\/p>\n
# umount \/mnt<\/em><\/p>\n
# mount -o subvol=volume1 \/dev\/sdb1 \/mnt<\/em><\/p><\/blockquote>\n
With these commands we can able to mount particular subvoumes to our mount points .<\/p>\n
# btrfs sub get-default \/mnt<\/em><\/p><\/blockquote>\n
Above command will show default sub volume .<\/p>\n
To create new sub volume under volume1<\/em> ,<\/p>\n
# btrfs sub create \/mnt\/newsub1<\/em><\/p>\n
# btrfs sub create \/mnt\/newsub2<\/em><\/p><\/blockquote>\n
Above command will create new subvolume newsub1<\/em> under volume1<\/em> .<\/p>\n
Now the subvolume tree is look like,<\/p>\n
VOLUME1 -> NEWSUB1<\/em><\/strong><\/p>\n
-> NEWSUB2<\/em><\/strong><\/p>\n
Ok. Now we need to create another main subvolume like volume1<\/em> . For that ,<\/p>\n
# umount \/mnt<\/em><\/p>\n
# mount -o subvolid=0 \/dev\/sdb1 \/mnt<\/em><\/p>\n
# btrfs sub create \/mnt\/volume2<\/em><\/p>\n
# cd \/mnt<\/em><\/p>\n
# ls<\/em><\/p>\n
volume1\u00a0\u00a0\u00a0 volume2<\/em><\/p><\/blockquote>\n
<\/p>\n
Snapshot using BTRFS<\/strong><\/p>\n
Now we know how to create btrfs filesystem and subvolumes . This is only\u00a0 just beginning , using BTRFS we can configure RAID levels , remote transfer..etc . Later we can discuss about these mechanism .<\/p>\n
How can we take a snapshot ?\u00a0 Its simple .<\/p>\n
Consider we need to take a snapshot of volume2 .<\/em><\/p>\n
# cd \/mnt<\/p>\n
# btrfs sub snap volume2 volume2.snap<\/p><\/blockquote>\n
Format is btrfs sub snap <source> <destination><\/em><\/p>\n
# btrfs sub list \/mnt<\/p><\/blockquote>\n
This command will list all sub volumes including snapshots .<\/p>\n
<\/p>\n
How to take CentOS 7 root (\/) snapshot ?<\/strong><\/p>\n
By default CentOS\/RHEL is installing using XFS file system . Try to install centos using BTRFS file system . We can convert our existing file system to BTRFS , that will cover on next section .<\/p>\n
After installing CentOS\/rhel using BTRFS filesystem ,<\/p>\n
# btrfs sub list \/<\/p>\n
ID 257 gen 871 top level 5 path root
\nID 260 gen 41 top level 257 path var\/lib\/machines<\/p><\/blockquote>\nFrom this we know that root and var\/lib\/machines are the sub volumes . But from below command we can see that default sub volume mounted under \/ is root (ID 5). For more clarification you can check \/etc\/fstab<\/em> .<\/p>\n
# btrfs sub get-default \/<\/p>\n
ID 5 (FS_TREE)<\/p><\/blockquote>\n
To take snapshot we need to mount btrfs partition to other folder , for that first we need to know which partition .<\/p>\n
# df -h<\/p>\n
Filesystem\u00a0\u00a0\u00a0\u00a0\u00a0 Size\u00a0 Used Avail Use% Mounted on
\n\/dev\/sda3\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 18G\u00a0 1.1G\u00a0\u00a0 15G\u00a0\u00a0 7% \/<\/p><\/blockquote>\nFrom this we can see that \/dev\/sda3<\/em> is the partition which we need to mount .<\/p>\n
# mkdir \/btrfs<\/p>\n
# mount -o subvolid=0 \/dev\/sda3 \/btrfs<\/p>\n
# cd \/btrfs<\/p>\n
# ls<\/p>\n
root<\/p><\/blockquote>\n
To make snapshot,<\/p>\n
# btrfs sub snap root root-$(date +%Y%m%d)<\/p>\n
# ls<\/p>\n
root\u00a0 root.snap<\/span><\/p>\n
# btrfs sub list \/<\/p>\n
ID 257 gen 946 top level 5 path root
\nID 260 gen 41 top level 257 path root\/var\/lib\/machines
\nID 262<\/strong> gen 946 top level 5 path root-20160607<\/p><\/blockquote>\nNow we have one snapshot of our root file system .<\/p>\n
To boot with our new snapshot need to set some configurations ,<\/p>\n
Remove all the \u201crootflags=subvol=root\u201d arguments from\u00a0\/boot\/grub2\/grub.cfg<\/span><\/em>. If you don\u2019t do this, it will disregard the default subvolume id we just set and always boot into the root subvolume.<\/p>\n
# sed -i \u2018s\/rootflags=subvol=root \/\/\u2019\u00a0\/boot\/grub2\/grub.cfg<\/span><\/p>\n
# btrfs sub set-default 262 \/.<\/p>\n
# reboot<\/p><\/blockquote>\n
After booting up , we need to check root subvolume .<\/p>\n
# btrfs subvolume show \/.<\/p>\n
Name:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 root-20160607
\nuuid:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 1a65c55f-4e07-134c-8e34-f17574d2f4ac
\nParent uuid:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 dfd7589a-0fa4-5c4b-8b95-312d6c619f69
\nCreation time:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 2016-06-07 02:05:05
\nObject ID:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 262
\nGeneration (Gen):\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 958
\nGen at creation:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 946
\nParent:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 5
\nTop Level:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 5
\nFlags:\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 \u2013
\nSnapshot(s):<\/p><\/blockquote>\nSuccess . Now we know how to create snapshot and point that to default right ?<\/p>\n
So , how can we delete those unwanted subvolumes ?<\/strong> Its easy .<\/p>\n
# mount -o subvolid=0 \/dev\/sda3 \/btrfs<\/p>\n
# cd \/btrfs<\/p>\n
# ls<\/p>\n
root root-20160607<\/p>\n
# btrfs sub delete root<\/p><\/blockquote>\n
Default rm<\/em> command will not work here .<\/p>\n
<\/p>\n
Snapper Utility<\/strong><\/p>\n
Snapper is an advanced tool for snapshoting BTRFS systems\u00a0 .<\/p>\n
To install ,<\/p>\n
# yum install snapper -y<\/p><\/blockquote>\n
Snapper utility is working with configuration\u00a0 file and .snapshot<\/em> hidden directory . Configurations are stored under \/etc\/snapper\/configs\/<\/em> location.<\/p>\n
When we are run a snapper configuration it will automatically create conf file under \/etc<\/em> and create .snapshot directory under that sub volume .<\/p>\n
For example ,<\/p>\n
We can use \/dev\/sdb1<\/em> and volume1<\/em> as default sub volume .<\/p>\n
# mount -o subvol=volume1 \/dev\/sdb1 \/mnt<\/p>\n
# cd \/mnt<\/p>\n
# snapper -c volume1 create-config \/mnt\/<\/p><\/blockquote>\n
This command will create volume1<\/em> config file under \/etc\/snapper<\/em> and .snapshot<\/em> folder on this location .<\/p>\n
To take a snapshot using snapper ,<\/p>\n
# snapper -c volume1 create -d \u201cFirst snapshot\u201d<\/p>\n
# snapper -c volume1 list<\/p><\/blockquote>\n
To see the difference between snapshots ,<\/p>\n
# snapper -c volume1 status 1..0<\/p><\/blockquote>\n
This means snapper is comparing snapshot ID 1 with 0 and list differences .<\/p>\n
If you need to recover previous snapshot files ,<\/p>\n
# snapper -c volume1 undochange 1..0<\/p><\/blockquote>\n
Task to do your self<\/strong> <\/em>: Create some files under sub volume and take snapshot . Revert back using snapper.<\/p>\n","protected":false},"excerpt":{"rendered":"
In these days we can find lot of file systems and they are also used in many Operating systems and block devices . Comparing with this file systems BTRFS is a newbie . Btrfs is a new copy on write (CoW) filesystem for Linux aimed at implementing advanced features while focusing on fault tolerance, repair […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[73],"tags":[],"_links":{"self":[{"href":"https:\/\/mohan.sg\/index.php?rest_route=\/wp\/v2\/posts\/6608"}],"collection":[{"href":"https:\/\/mohan.sg\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/mohan.sg\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/mohan.sg\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/mohan.sg\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=6608"}],"version-history":[{"count":1,"href":"https:\/\/mohan.sg\/index.php?rest_route=\/wp\/v2\/posts\/6608\/revisions"}],"predecessor-version":[{"id":6609,"href":"https:\/\/mohan.sg\/index.php?rest_route=\/wp\/v2\/posts\/6608\/revisions\/6609"}],"wp:attachment":[{"href":"https:\/\/mohan.sg\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=6608"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/mohan.sg\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=6608"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/mohan.sg\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=6608"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}