osdir.com

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Options to replace hardware of the cluster


Both are good options, a new data center in same cluster will work faster as you can stream multiple nodes.
Or you can add 8 nodes to existing data center and then decommission one node at a time but this is lengthy process including cleanup after decommission.
One node at a time is equally good. If data isn’t much this step can be efficient.

One approach we did was:
RSync a temporary storage volume online
Stop c*
Rsync delta
Detach temp storage
Start a new node with new hardware
Attach volume
Rsync to actual storage

Repeat process on each node one at a time


Sent from my iPhone

On Jun 14, 2018, at 4:10 AM, Christian Lorenz <Christian.Lorenz@xxxxxxxxxxxx> wrote:

Hi,

 

we need to move our existing cassandra cluster to new hardware nodes. Currently the cluster size is 8 members, they need to be moved to 8 new machines. Cassandra version in use is 3.11.1.  Unfortunately we use materialized views in production. I know that they have been marked retroactively as experimental.

What is a good way to move to the new machines? One-by-One, or setup a new cluster as a separate DC? The move should be done without downtime of the application.

 

Do you have some advice for this kind of maintenance task?

 

Kind regards,

Christian