Re: [Hampshire] LVM mirroring without software RAID

Top Page
Author: Adam Trickett
Date:  
To: Hampshire LUG Discussion List
CC: 
Subject: Re: [Hampshire] LVM mirroring without software RAID

Reply to this message
gpg: failed to create temporary file '/var/lib/lurker/.#lk0x56d54100.hantslug.org.uk.15523': Permission denied
gpg: keyblock resource '/var/lib/lurker/pubring.gpg': Permission denied
gpg: Signature made Fri Mar 14 17:22:10 2008 GMT
gpg: using DSA key 019AD0D8166C4BF0
gpg: Can't check signature: No public key
On Friday 14 Mar 2008, Adrian Bridgett wrote:
> On Thu, Mar 13, 2008 at 21:53:09 +0000 (+0000), Adam Trickett wrote:
> > Hi,
> >
> > Linux's LVM implementation is from HP-UX and doesn't do mirroring for
> > that you need another application. IBM's original LVM solution can do
> > mirroring of a within a volume group, so you don't need a RAID
> > application first.
>
> IIRC it's the syntax which comes from HP-UX, not the code or
> implementation. Wikipedia says the "design was based on HP-UX".


Fair enough, though I will admit that the HP-UX syntax is not logical, the AIX
application naming is much more consistent with other Unix commands.

> Yes, IBM's (well, AIX - IBM's Unix) does upto 3-way mirroring as part
> of the LVM. IIRC (it's been a few years now!) each LV could have upto 1023
> LPs (logical paritions), each of those referred to 1, 2 or 3 PPs
> (physical paritions). I'm pretty sure these limits were busted
> properly with AIX5 (there were hacks in AIX4 ("-f factor" IIRC)).
>
> > Apparently it's now possible to do this in Linux, from Debian Etch and
> > Centos 4 era systems. However I've not found out how, almost everyone
> > talks about running LVM2 on top of standard Linux software RAID.
>
> When I looked into it, I decided to stick with LVM and MD - partly the
> lack of documentation made me believe that it wasn't the robust choice.


The lack of documentation is a bit scary. I'll try it out in a VM and see what
I find.

--
Adam Trickett
Overton, HANTS, UK

I've found that people who are great at something are
not so much convinced of their own greatness as
mystified at why everyone else seems so incompetent.
    -- Paul Graham, "Great Hackers" 2004