You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@cassandra.apache.org by onmstester onmstester <on...@zoho.com.INVALID> on 2021/02/15 06:25:54 UTC
number of racks in a deployment with VMs
Hi,
In a article by thelastpickle [1], i noticed:
The key here is to configure the cluster so that for a given datacenter the number of racks is the same as the replication factor.
When using virtual machines as Cassandra nodes we have to set up the cluster in a way that number of racks is the same as physical servers, so by losing one physical server just one copy of any data would be lost, right? which could be much greater than RF, would this cause any harm on cluster health/balance/availability?
[1]: https://thelastpickle.com/blog/2021/01/29/impacts-of-changing-the-number-of-vnodes.html
Sent using https://www.zoho.com/mail/
Re: number of racks in a deployment with VMs
Posted by Bowen Song <bo...@bso.ng.INVALID>.
The reason for the zoho (and my) emails go to the spam box is because
the Apache mailing list software is messing around with the DKIM
signature and the "From:" address. I have created INFRA-21415
<https://issues.apache.org/jira/browse/INFRA-21415> for this.
On 15/02/2021 22:36, Kane Wilson wrote:
> There are operational advantages to having #racks == RF, however it's
> by no means mandatory. Having more racks than RF doesn't cause any
> availability/health/balance problems, it is only disadvantageous in
> that it makes some cluster maintenance tasks more expensive/unwieldy
> like repairs and DC migrations. Note that you can re-use racks across
> your physical servers assuming you have more physical servers than RF,
> it's not necessary to have a separate rack per physical server, as
> long as there are at least RF physical servers (ideally these physical
> servers should also be in the same rack/AZ in the datacenter to
> protect against physical rack/datacenter outage).
>
> FYI this went to spam on GMail. I suspect google is flexing their
> monopoly muscles against zoho.
>
> On Mon, Feb 15, 2021 at 5:26 PM onmstester onmstester
> <on...@zoho.com.invalid> wrote:
>
> Hi,
> In a article by thelastpickle [1], i noticed:
>
> The key here is to configure the cluster so that for a given
> datacenter the number of racks is the same as the replication
> factor.
>
>
> When using virtual machines as Cassandra nodes we have to set up
> the cluster in a way that number of racks is the same as physical
> servers, so by losing one physical server just one copy of any
> data would be lost, right? which could be much greater than RF,
> would this cause any harm on cluster health/balance/availability?
>
> [1]:
> https://thelastpickle.com/blog/2021/01/29/impacts-of-changing-the-number-of-vnodes.html
> <https://thelastpickle.com/blog/2021/01/29/impacts-of-changing-the-number-of-vnodes.html>
>
>
> Sent using Zoho Mail <https://www.zoho.com/mail/>
>
>
>
Re: number of racks in a deployment with VMs
Posted by Kane Wilson <k...@raft.so>.
There are operational advantages to having #racks == RF, however it's by no
means mandatory. Having more racks than RF doesn't cause any
availability/health/balance problems, it is only disadvantageous in that it
makes some cluster maintenance tasks more expensive/unwieldy like repairs
and DC migrations. Note that you can re-use racks across your physical
servers assuming you have more physical servers than RF, it's not necessary
to have a separate rack per physical server, as long as there are at least
RF physical servers (ideally these physical servers should also be in the
same rack/AZ in the datacenter to protect against physical rack/datacenter
outage).
FYI this went to spam on GMail. I suspect google is flexing their monopoly
muscles against zoho.
On Mon, Feb 15, 2021 at 5:26 PM onmstester onmstester
<on...@zoho.com.invalid> wrote:
> Hi,
> In a article by thelastpickle [1], i noticed:
>
> The key here is to configure the cluster so that for a given datacenter
> the number of racks is the same as the replication factor.
>
>
> When using virtual machines as Cassandra nodes we have to set up the
> cluster in a way that number of racks is the same as physical servers, so
> by losing one physical server just one copy of any data would be lost,
> right? which could be much greater than RF, would this cause any harm on
> cluster health/balance/availability?
>
> [1]:
> https://thelastpickle.com/blog/2021/01/29/impacts-of-changing-the-number-of-vnodes.html
>
> Sent using Zoho Mail <https://www.zoho.com/mail/>
>
>
>