[NBLUG/talk] RAID1 and partitions

Omar Eljumaily omar at omnicode.com
Fri May 1 12:33:10 PDT 2015


Actually, the AWS config I'm thinking about would get archived much more 
often than every night, at least every 15 minutes.

On 5/1/2015 12:23 PM, Omar Eljumaily wrote:
> I don't like RAID either for a number of reasons.  However, the 
> clients I deal with typically are losing about $500 per hour if one of 
> their main servers goes down.  I've actually relied on RAID a number 
> of times.  It's saved everybody lots of money and head aches.
>
> I'd like to go full cloud like on AWS or something, but it's too slow 
> and sometimes too expensive right now.  As an alternative, I'm 
> shooting for a main server(s) locally that get archived every night on 
> AWS with the ability to work completely off the cloud if the local 
> server goes down.
>
> I think the local server will still use RAID.  It's just a matter of 
> practicality.  1 TB of data takes multiple hours to restore, and 
> non-active mirrors are marginally reliable at best.
>
> Omar
>
>
> On 5/1/2015 12:10 PM, gandalf at sonic.net wrote:
>> I'm not really a big fan of raid. I think your much better off having 
>> a second drive and doing a daily rsync. Raid drives typically fail 
>> within the same timeframe, also it is not uncommon for a drive to 
>> fail in a manner that it corrupts the other drive. Regular backups 
>> are much more powerful.
>>
>> There I said it. Everyone chime in and tell me I'm an idiot. :-)
>>
>>
>> On 2015-05-01 10:23, Omar Eljumaily wrote:
>>> Does anybody have any tips for setting up a RAID1 array on Ubuntu?
>>>
>>> I've looked at this:
>>>
>>> https://help.ubuntu.com/lts/serverguide/advanced-installation.html
>>>
>>> It suggests setting up
>>>
>>> md0 as a swap partition and
>>> md1 as an ext4 / partition which is also bootable.
>>>
>>> My experience with Centos is to do:
>>>
>>> md0: /boot with ext2
>>>
>>> md1: as an LVM volume creating partitions:
>>>
>>> swap
>>> and / as ext4
>>>
>>> My goal is reliability and to be able to boot into either volume when
>>> the other is degraded.  Is there any standard way to do this?
>>>
>>> Thanks,
>>>
>>> Omar
>>>
>>>
>>>
>>> _______________________________________________
>>> talk mailing list
>>> talk at nblug.org
>>> http://nblug.org/cgi-bin/mailman/listinfo/talk
>> _______________________________________________
>> talk mailing list
>> talk at nblug.org
>> http://nblug.org/cgi-bin/mailman/listinfo/talk
>
> _______________________________________________
> talk mailing list
> talk at nblug.org
> http://nblug.org/cgi-bin/mailman/listinfo/talk



More information about the talk mailing list