1.0

Table Of Contents
Site Recovery Manager Administration Guide
74 VMware, Inc.
To prepare Site B for failover in case of a second disaster
1AtSiteB,rescanthehostbusadapters(HBAs)onallthehosts.Thismakesiteasier
toidentifytheprotectedvirtualmachines,becausetheynowappearinvalidinthe
inventory,asdoestheprotectiongrouptowhichtheyareassigned.
2AtSiteB,removePG2fromtherecoveryplan,then
deletePG2.
3AtSiteB,removealltheprotectedvirtualmachinesthatwererecoveredtoSite A
frominventoryatSiteB.Inthisscenario,thisisapp_vm7 to app_vm12.(See“To
removetheoutofdateprotectedvirtualmachinesfrominventoryonSiteA.)
4 CreatePG3atSiteAfor
theprotectedvirtualmachines.
PG3shouldbeidenticaltoPG 1,theprotectiongroupthatwasoriginally
associatedwithRP1,therecoveryplanthatwasruninRecoverymodeand
resultedinthestartupoftheprotectedvirtualmachinesatSite B.
5ReassociatetheprotectiongroupsatSiteAwithRP
1atSiteB.
YoudonotneedtodeleteRP2(therecoveryplanthatwascreatedatSiteAto
facilitatetherecoverybacktoSiteAfromSiteB).
YouhavecompletedthereprotectionoftheprotectedvirtualmachinesatSiteA.
VMwarerecommendsthatyounowconduct
afailovertestagainstRP1toensurethat
SiteAisprotectedandreadyforanyeventthatmaynecessitatearecoverytoSiteBif
anotherdisasteroccurs.
Other Failback Considerations
Thereareseveralmoreissuestoconsiderduringafailback.
SitePairing—SitesAandBneedtobepairedonlyonce.SRMmaintainsa
bidirectionalrelationshipbetweenpairedsites.
DNSUpdates—IfSiteAandSiteBarenotjoinedbyastretchedVLAN,manually
provideDNSupdatesasthevirtualmachinesaremovedbetweenSite AandSite B
andtheirIPaddresseschangetoaccommodatetheirnewnetwork.