introduction
in this article we will see how to configure HA and verify it by power off ESXI153 to see what will happen
so please back to part I of HA to understand the concept before configure HA effectively
HA network Diagram
to beter understand HA situation :
please look to diagram above
we have clluster16 contain three ESXI host : ESXI151, ESXI152 ,and ESXI153
all three ESXI server are connected to local storage beside of shared storage [NFS Datastore, and ISCSI datastore ],
and also belong the same management subnet 172.16.x.x/16 , as well as same storage sunbet 172.21.x.x/16
also we have VM windows 10 that run by ESXI153 on ISCSI datastore
as you see below
HA configuration
before deep diving in HA configuration : please make sure your vSphere environment meet listed HA requirement in part 1 here
now to configure HA : open vcenter and select cluster > configure > enable HA
monitor HA configuration
it’s better to monitor your vsphere HA from time to time to see what is going on
open vcenter > cluster > monitor >vsphere HA > summary
please note that ESXI153 is selected randomly as HA master and other ESXI151 + ESXI152 as slave
also we have 2 VM is protected : VCSA171 and Windows10 VM
before ESXI153 down accidentally
before ESXI153 down accidentally 🙂 , please have look to vsphere environment
all esxi server is up and online
also VM windows 10 is running on ISCSI datastore by ESXI153
after ESXI153 down accidentally
let us now to suppose that ESXI153 accidently down for any reason [ power outage for example ]
after HA heart beating [5 minute ] : HA will discover that ESXI153 is down >
SO ….
it will ask any free server [here is ESXI152] to run VMs that was running by ESXI153 before down 5
after ESXI153 back online
now HA are able to manage life in absence of ESXI153
what will happen if ESXI153 back online again [we fixed power outage ]
vcenter will detect three ESXI online
and windows 10 VM will keep running by ESXI152 , until we decide to migrate it manually to ESXI153
Conclusion
vsphere HA is great feature to protect VM in your environment
BUT
there is one disadvantage : we have down time 3 minute and 2 minutes the time to VM itself on another ESXI
[ approximately 3+ 2 = 5 minute ]
this is called RTO Recovery time Object
some VM accept down 5 minutes
BUT
some critical VMs does not accept that , even it doesn’t accept a few seconds of down time
so we have to use another technique called [Fault Tolerance ] which make downtime = 0 [yes ZERO]
Next article in HA part 3 , we will discuss HA advanced feature and recommendation
before moving to Fault Tolerance
I am maher islaieh thanking you for join us
