One obvious issue is disk failure toleration . As in if RF =3 on.normal JBOD disk failure toleration is 2. In RAID 5 one can loose more than only one disk RAID here will be data corruption. effectively making the broker unusable, thus reducing our drive failure toleration to 2 drives ON 2 different brokers with the added caveat that we loose the whole broker as well ?
On Mon, Mar 23, 2020 at 10:42 PM Vishal Santoshi <vishal.santo...@gmail.com> wrote: > We have a pretty busy kafka cluster with SSD and plain JBOD. We > planning or thinking of using RAID 5 ( hardware raid or 6 drive SSD > bokers ) instead of JBID for various reasons. Hss some one used RAID 5 ( we > know that there is a write overhead parity bit on blocks and recreating a > dead drive ) and can share there experience on it . Confluent advises > against it but there are obvious ease one gets with RAID ( RAID 10 is to > expensive space wise ) .... Any advise /comments etc will be highly > appreciated. > > Regards. > >
