loopback interface是系统虚拟出来的一个接口。所以不会因为线路或接口问题而down掉
loopback interface一般是绑定的127.0.0.1回环地址(只用于本机系统内的地址)
loopback interface也可以绑定其他类型的ip，比如局域网ip或者公网ip，用于实现特殊目的
比如LVS的实现的后台服务器上（多个机器的loopback都设置为一个同样的ip）都可以对这个ip地址提供处理服务，而LVS的前端根据一定的分配策略将dest是这个ip的报文分发到这些机器上以后进行处理
在比如：bgp的peer对等体互联实现中多建议将 可达ip地址放在loopback interface上。并建议用此ip生成BGP Router ID
there has one simular project "https://github.com/doczjs/docz" like docusaurus.
day after day, too
I saw you standing there
I wished you in my shows
could you be there
could we stay together babe
cause this love knocks off my way
uh la la lala
uh la la lala
uh la la lala
1, 2, 3 babe
uh la la lala
uh la la lala
1, 2, 3 is the sign for you
to tell me that you love me
Tell me the reason
why this should be true
I never heard you say
three words I love you
How could you show me
this world`s been broken too
take the chance to rescue me
could we stay together babe
cause this love knocks off my way
uh la la lala
uh la la lala
uh la la lala
1, 2, 3 babe
uh la la lala
uh la la lala
1, 2, 3 is the sign for you
to tell me that you love me
1, 2, 3 I will count up to
don`t run away cause I gonna get you
1, 2, 3 is the sign for you
to tell me that you love me
uh la la lala
uh la la lala
uh la la lala
1, 2, 3 babe
uh la la lala
uh la la lala
1, 2, 3 is the sign for you
to tell me that you love me
假设磁盘数量为n,单磁盘读写性能为x
| raid | 读速度 | 写速度 | 冗余 | 实现原理 |
|---|---|---|---|---|
| raid0 | n*x | x | 无 | Stripe/条带化/分块分布/多盘存储，以提高吞吐速度，无冗余备份 |
| raid1 | (n/2)*x | x | n/2备/2 | Mirror 一个数据分开写两份 |
| raid5 | (n-1)*x | x/2(写操作同时需要带有读取和计算操作性能差) | n-1备1 | 分布式奇偶校验,n-1个份的数据,剩下1份是前n-1份的奇偶校验XOR汇总,用于n-1有一份丢失后的恢复计算,商业上用的最多 |
| RAID6 |
mdadm --create /dev/md0 --level=1 --raid-devices=2 /dev/sdc2 /dev/sda2
mdadm --create /dev/md1 --level=10 --raid-devices=4 /dev/sd[b-e]1
mkfs.ext4 /dev/md0
e2label /dev/md0 raid1t
mount /dev/md0 /mnt/raid1t/
mdadm --detail --scan --verbose >> /etc/mdadm/mdadm.conf
mdadm --assemble --scan
cat /proc/mdstat
cat /sys/block/md0/md/sync_action
mdadm --detail --scan --verbose
mdadm --detail /dev/md0 #check array
mdadm --examine /dev/sd[a-e] #check array item(component)
mdadm --examine /dev/sd[a-e]1
mdadm /dev/md0 --add /dev/sdc2
mdadm /dev/md0 --add /dev/sde2
mdadm /dev/md0 --fail /dev/sda2
mdadm /dev/md0 --remove /dev/sda2
mdadm /dev/md0 --add /dev/sda2
mdadm /dev/md0 --add /dev/sdd2
mdadm /dev/md0 --remove /dev/sdd2
mdadm --stop /dev/md0
mdadm --remove /dev/md0
mdadm --zero-superblock /dev/sdf1 /dev/sde1
if you not save mdadm to mdadm.conf, or move disk to other new machin mdadm --detail --scan --verbose, will name raid as /dev/md127 (external device start number) better try fix name /dev/md127 to /dev/md0 by: ref about md127
mdadm --assemble --scan
mdadm --detail --scan --verbose
mdadm --detail --scan --verbose >> /etc/mdadm/mdadm.conf
vi /etc/mdadm/mdadm.conf #modify md127 to md0,
update-initramfs -u
reboot
是否有跨网raid1？(基于iscsi等)
| btrfs | ||
| zfs | ||
| zfs |
Welcome to danfestar blog
This blog is power by https://docusaurus.io and plugin plugin-content-blog
I read about Little Newton's Theorem in a AMC8 training book
It‘s attractive
But rare topic discuss on this Theorem
Here are few article listed: