[Linux-cluster] Two node failover with shared storage and Luns in active/passive

[Rick Stevens]

On 08/29/2016 06:23 AM, Amjad Syed wrote:
> Hello,
>  We have implemented a two node failover cluster using RHEL pcs in
> active passive mode
> We have shared storage with one LUN,
> Our concern is what if this LUN fails or there is hardware problem
> accessing the data in it.
> What is the best  way to synchronize data between two LUNS ? any
> suggestion will help.

If you have two identical LUNs, you could run something like drbd to
mirror one LUN to a backup:

	https://www.drbd.org/en/comp/drbd-linux-driver

Note this is done at the block level. It doesn't look at filesystem-
specific stuff, it just watches block-level disk writes to the primary
device and replicates those to the backup over the network.

There are other things you can do at a filesystem level, such as
rsyncing often, corosync, etc.

----------------------------------------------------------------------
- Rick Stevens, Systems Engineer, AllDigital    ricks at alldigital.com -
- AIM/Skype: therps2        ICQ: 226437340           Yahoo: origrps2 -
-                                                                    -
-     Squawk!  Pieces of Seven!  Pieces of Seven!  Parity Error!     -
----------------------------------------------------------------------