Tom Kunz wrote:
>
<snip>
> Linux-based software alternative to the super-expensive external RAID
> towers that have multiple independent SCSI buses. They run for $10k
> each, and you can connect multiple machines into them, which will all
> mount the array simultaneously. Any node can go down at any time,
> regardless of any cron schedule, and no data will be lost.
<snip>
That will do _nothing_ for you, because:
1.) you can only mount it r/w on exactly one machine.
2.) even if 1) is ok for you, you cannot even mount the array ro on the
other machines, because of Linux' disk caching.
Maybe raid1 over NBD (see linux/Documentation/nbd.txt) is what you want,
but I don't know if that works.
Basically, what I think of is the following:
- Machine A is master and has one half of the disk, raid1'ed with the
nbd'ed other half from Machine B, which is the slave.
- If Machine A fails, B detects this somehow, restarts its half of the
raid1 disk in degraded mode, mounts it r/w and takes the place of A.
If then A comes up again, it treats its half of the array faulty and
reconstructs it after the nbd'ed half of B.
- If Machine B goes down, Machine A's raid 1 falls to degraded mode
until B comes up again, at which point 's ndb half is reconstructed
after A's half.
If this works, you can also add a third machine and make a threefold
raid1 for added HA. Curious myself if this would work. Unfortunately
cannot test this myself.
Marc
--
Marc Mutz <[EMAIL PROTECTED]> http://marc.mutz.com/
University of Bielefeld, Dep. of Mathematics / Dep. of Physics
PGP-keyID's: 0xd46ce9ab (RSA), 0x7ae55b9e (DSS/DH)
