VMware Cloud Community
nblr06
Enthusiast
Enthusiast
Jump to solution

how will vsan disk failure event impact on VM's filesystem?

just curious about this question, assume that we are talking about vSAN version 6.7 or 7.

If a VM(windows or linux) lives in a vSAN storage with policy ftt=1 and one of the data copy of this VM encounters vSAN disks permanent failure in which the copy resides in, will this situation influence VM's filesystem???

According to the mechanism of vSAN data redundancy, I doubt that such kind of accident will damage the VM operating system's file systems but not 100% sure.

any ideas?

0 Kudos
1 Solution

Accepted Solutions
fabio1975
Commander
Commander
Jump to solution

Ciao 

Premise:

------

vSAN uses the Fault Domains concept to group hosts into pools. Each FD can have one or more ESXi hosts.
It is typically used to protect the cluster from rack or site failure.
vSAN will never place components of the same object in the same FD. If the entire FD fails (top of the rack switch failure, site disconnection), we will still have the majority of votes for the item to be available.
If we don't configure any FD in vCenter, each ESXi host will become a kind of FD, because we will never have components of the same object on the same host ... even if the host has more than one disk group.
So the smallest FD is the host itself 

-----

I suppose that in your situation there is no FDs configured and therefore the single ESXi HOST is an FD.
What happens to a VM if one of the vSAN ESXi hosts fails depends on the Storage Policy applied to the VMDK of the VM and on which ESXi host the VM resides.


In the configuration of the storage policies we have various possibilities of data redundancy:

fabio1975_0-1639871789216.png

 


If you select NO data Redundancy .... you have no protection of the VMDK with which you have associated the policy (and I do not recommend using it)
If, on the other hand, you select one of the other policies, you have a VMDK protection (With various levels of fault protection)

Possible scenarios of what happens to VMs (and their disks) can be:

fabio1975_1-1639871937441.png

 

Fabio

Visit vmvirtual.blog
If you're satisfied give me a kudos

View solution in original post

3 Replies
fabio1975
Commander
Commander
Jump to solution

Ciao

Normally FTT = 1 (Number of Failures to Tolerate) must be accompanied by a RAID policy (1 or 5/6 apart from the configuration of the stretched cluster).

Therefore the fault of a disk (and of the respective disk group) does not affect the functioning of the VMs that have a copy of the disk on that diskgroup.

Fabio

Visit vmvirtual.blog
If you're satisfied give me a kudos

0 Kudos
nblr06
Enthusiast
Enthusiast
Jump to solution

@fabio1975 

thank you so much! 

I got last question about vSAN:

will ESXi failure event(such as the disk that ESXi installed in is damaged) cause VM(that runs in vSAN storage) data lose or some kind of malfunctioning ?

0 Kudos
fabio1975
Commander
Commander
Jump to solution

Ciao 

Premise:

------

vSAN uses the Fault Domains concept to group hosts into pools. Each FD can have one or more ESXi hosts.
It is typically used to protect the cluster from rack or site failure.
vSAN will never place components of the same object in the same FD. If the entire FD fails (top of the rack switch failure, site disconnection), we will still have the majority of votes for the item to be available.
If we don't configure any FD in vCenter, each ESXi host will become a kind of FD, because we will never have components of the same object on the same host ... even if the host has more than one disk group.
So the smallest FD is the host itself 

-----

I suppose that in your situation there is no FDs configured and therefore the single ESXi HOST is an FD.
What happens to a VM if one of the vSAN ESXi hosts fails depends on the Storage Policy applied to the VMDK of the VM and on which ESXi host the VM resides.


In the configuration of the storage policies we have various possibilities of data redundancy:

fabio1975_0-1639871789216.png

 


If you select NO data Redundancy .... you have no protection of the VMDK with which you have associated the policy (and I do not recommend using it)
If, on the other hand, you select one of the other policies, you have a VMDK protection (With various levels of fault protection)

Possible scenarios of what happens to VMs (and their disks) can be:

fabio1975_1-1639871937441.png

 

Fabio

Visit vmvirtual.blog
If you're satisfied give me a kudos