VMware Cloud Community
suhag79
Hot Shot
Hot Shot
Jump to solution

slack space for 3-Node vSAN cluster

Hi all,

We are going to setup 3 Node vSAN cluster with essential plus licensing. All three nodes will be installed in remote office and vCenter essential appliance will be part of the same 3 Node vSAN cluster. 

Total 15 Vms are going to be host and storage requirement for these VMs are 10 TB. We will be setting up RAID-1 and FTT-1 policy so storage requirement will be 10 TB*2 = 20 TB. I want to know, since this is 3 node vSAN cluster, should i consider 30% slack space for the component rebuild ? i guess No ?

Thanks,

1 Solution

Accepted Solutions
TheBobkin
Champion
Champion
Jump to solution

Hello suhag79

"I want to know, since this is 3 node vSAN cluster, should i consider 30% slack space for the component rebuild ? i guess No ?"

Always consider slack space - even if you are never planning on changing Storage Policies in such a way that they will need a lot of resync space (e.g. changing the Stripe Width), you will inevitably require some space for resyncs following maintenance/upgrades/updates of single nodes.

The more important thing to consider here is slack space to recover from a single disk/Disk-Group failure - if you have no space to rebuild data on this node then your data is FTT=0 and you also won't be able create new FTT=1 Objects until you get this fixed.

Additionally allowing some slack space for growth and expansion is generally a better option than waiting until you are out of space and facing the issues that come with this to add more storage.

Bob

View solution in original post

3 Replies
suhag79
Hot Shot
Hot Shot
Jump to solution

Hello all,

Can anyone help on this. Should we consider slack space in storage sizing for 3 Node vSAN cluster ?

Thanks,

Reply
0 Kudos
TheBobkin
Champion
Champion
Jump to solution

Hello suhag79

"I want to know, since this is 3 node vSAN cluster, should i consider 30% slack space for the component rebuild ? i guess No ?"

Always consider slack space - even if you are never planning on changing Storage Policies in such a way that they will need a lot of resync space (e.g. changing the Stripe Width), you will inevitably require some space for resyncs following maintenance/upgrades/updates of single nodes.

The more important thing to consider here is slack space to recover from a single disk/Disk-Group failure - if you have no space to rebuild data on this node then your data is FTT=0 and you also won't be able create new FTT=1 Objects until you get this fixed.

Additionally allowing some slack space for growth and expansion is generally a better option than waiting until you are out of space and facing the issues that come with this to add more storage.

Bob

suhag79
Hot Shot
Hot Shot
Jump to solution

Thanks TheBobkin

Appreciate  the info.

Reply
0 Kudos