backup by using dd

Issues related to configuring your network
Post Reply
jos
Posts: 1
Joined: 2018/05/02 19:36:44

backup by using dd

Post by jos » 2018/05/02 20:08:15

hi,
I rented a remote server in a host firm. I installed centos 6, My remote server has 3 hard drives. I want to backup my all data in server by using dd command. One of the hard drives' capacity is 2 tb. To backup my all data in remote server require to run dd command approximately 60 hour uninterruptedly. But in average 15 hour, the network drops and dd imaging process interrupt. I want host firm support help me to prevent interruption. He said to me sshd setting must be config again and it is about sshd timeout. He said to me some parameters must be increase such as "TCPKeepAlive-yes"-"ClientAliveInterval-30"-"ClientAliveCountMax-300". Are the values optimal? I am not sure. Because network dropped. What are the optimal values to backup my all data by using dd command. Because network shouldn't drop for 50 hours. I need your help
Thank you

tunk
Posts: 356
Joined: 2017/02/22 15:08:17

Re: backup by using dd

Post by tunk » 2018/05/03 21:45:13

Would rsync be an alternative?
Or set it up as a cron job.

User avatar
TrevorH
Forum Moderator
Posts: 23213
Joined: 2009/09/24 10:40:56
Location: Brighton, UK

Re: backup by using dd

Post by TrevorH » 2018/05/03 23:35:38

I suspect that you're doing something wrong somewhere. For a start, in 60 hours, which is 60*60*60=216000 seconds, assuming a write rate of say 100MB/s - might be a bit generous but not unachievable - that would mean that in that 60 hours you should be able to write 21,600,000MB = 21093GiB or 20.6TB. Even if you can only write at 50MB/s which any decent hard disk should manage, that should still be sufficient time to write 10TB. Yet you say your disk is 2TB so that's writing at 10MB/s which is just rubbish.

Something doesn't add up.

Also, why are you doing a bit for bit clone of the entire disk anyway? If you're backing up the entire disk like this then you are also backing up all the unused space on the filesystems on the other one. And all of the filesystems on there at that when your data is the thing you really want a backup of - everything else can be reinstalled/reconfigured.

If you just want an exact clone of the disk and you want to keep it constantly up to date then perhaps you should be looking at using RAID 1?
CentOS 5 died in March 2017 - migrate NOW!
Full time Geek, part time moderator. Use the FAQ Luke

Post Reply