Oracle Data Guard 12.2 - 18c - doag.org · • In Oracle Database 12c Release 2, a storage failure at the primary database will be handled by the Oracle Data Guard Asynchronous process.

Copyright©2015,Oracleand/oritsaffiliates.Allrightsreserved.|

OracleDataGuard12.2-18c

UweKirchhoffTechnicalArchitectACSGlobalDelivery


SafeHarborStatementThefollowingisintendedtooutlineourgeneralproductdirection.Itisintendedforinformationpurposesonly,andmaynotbeincorporatedintoanycontract.Itisnotacommitmenttodeliveranymaterial,code,orfunctionality,andshouldnotberelieduponinmakingpurchasingdecisions.Thedevelopment,release,andtimingofanyfeaturesorfunctionalitydescribedforOracle’sproductsremainsatthesolediscretionofOracle.

3


OracleDataGuardProposedEnhancements

Manageability

Protection

Performance

PlannedMaintenance

ClientFailover

1

2

3

4

5

4


ManageabilityEnhancements12candBeyond

5


OracleDataGuard-Manageability

• CreateDataGuardStandbyswithDatabaseCreationAssistant(DBCA)• RMANDUPLICATEforFarSync•  EnterpriseManagersupportforFarSynccreation•  EMCLIstandbydatabasecreation

6

StandbyCreation


DBCAStandbyCreation

• AttheStandbyserverinstalltheOracleSoftwareandstartalistener•  Executeasinglelinecommand

• ConfigureyournetworkTNSNAMESforDataGuard• BuildyourBrokerconfiguration.– OrjustaddthisnewstandbytoyourexistingBrokerconfiguration

7

Simpleandscriptable!

dbca -silent -createDuplicateDB –gdbName PROD.de.oracle.com –sid STBY -sysPassword oracle -primaryDBConnectionString linbar01:1521/PROD.de.oracle.com -createAsStandby –dbUniquename SPROD -customScripts $ORACLE_HOME/PostCR.sql


RMANandFarSync

• RMANwillbeabletocreateaFARSYNCInstanceusingDUPLICATE–  duplicate target database for farsync from active database; –  duplicate target database for farsync backup …

•  Samebasicsetupasthestandbyactiveduplicateprocessthen…

8

connect target sys@PROD; connect auxiliary sys@PROD1FS; run {allocate channel prmy1 type disk; allocate auxiliary channel stby1 type disk; duplicate target database for farsync from active database spfile parameter_value_convert (’PROD',’SPROD') set 'db_unique_name'=’SPROD';}


EnterpriseManagerCloudControl

•  YouwillalsobeabletocreateFarSyncInstancesusinganew,butfamiliarEMCCcreationWizard

9

FarSyncCreationandManagement

Copyright©2015,Oracleand/oritsaffiliates.Allrightsreserved.| 10

FarSyncCreationandManagementEnterpriseManagerCloudControl


EnterpriseManagerCloudControl

•  EnterpriseManagerhashadaStandbyCreationWizardformanyversions•  ItalsohasacommandlineinterfacecalledEMCLI– Today(<12.2)youcannotscriptstandbycreationthroughit.– Butnotforlong!

– NoticehowthetargetcanbeaRACorSingleInstancedatabase– RMANduplicatefromactivedatabaseisalsothedefault

11

Commandlinestandbycreation

emcli create_standby -source_db_target_name=“PROD” -source_db_target_type=“rac_database” -dest_oracle_sid=“PROD1_1” -spname=“PROD1” -use_broker



• RESTfulServicesInterface– Creating,MonitoringandManagingaDataGuardBrokerConfiguration

• OracleDataGuardforDataWarehouses-NoLoggingsupport•  SubsetStandbys• AutomaticallySynchronizePasswordFiles•  TDETablespaceOfflineConversion

12

EaseofUse


DataGuardandNoForceLogging

• No-Loggingoperationsonlygenerateenoughredototellthestandbythatsomeblocksarenowunrecoverable.– Repairingthemmeantrestoringthecompleteaffecteddatafiles.

• No-LoggingblockswillbetrackedbythePrimarydatabase•  StandbyscanthenbeeasilyrepairedusingRMAN– Validationandrepaironlyofblocksonstandbythatwereinvalidatedbynon-loggedoperations(suchasdirectloads)ontheprimary

– ThiswillquerythePrimaryforblocklocationsthatcontainnon-loggedblocksandrepairthestandbydatabase.

13

rman validate/recover .. nonlogged block


RemovingtheheadacheofPasswordfilemanagement

•  Today(<12.2)whenyouchangethepasswordofanyelevatedprivilegeuser– YoumustcopythepasswordfiletoallothernodesofthePrimaryANDallstandbys– Madealittlesimplerin12.1.0.1byasharedpasswordfileforRAC

•  InOracleDatabase12cRelease2– Changingelevatedprivilegedpasswordsontheprimaryautomaticallyupdatesallthestandbys’passwordfiles.

– Noneedtomanuallycopypasswordfiles!

• NewDGredoauthenticationprotocolthatusesSSLcertificatefor‘redo_transport_user’

14

Password,Password,who’sgottherightpassword?


TDETablespaceConversion

• RequiresanExportandImporttoencrypt• Cannotbedoneonlinesodataisunavailableduringtheprocess• OutageforthedatacanbemitigatedbyusingaTransientLogicalStandby– ConvertaPhysicalStandbytoaTransientLogicalStandby•  Exportthedata,Createtheencryptedtablespace,Importthedataatthetransientlogicalstandby

– Switchover– Athttp://www.oracle.com/goto/maaseethesetwopapers•  ConvertingtoTransparentDataEncryptionUsingActiveDataGuard12c(DBMS_ROLLING)•  ConvertingtoTransparentDataEncryptionUsingDataGuardTransientLogicalStandbywithOracleDatabase11g

15

<12.2


TDETablespaceConversion

• Offlinedatafilemethodstillavailable– CompletewithPhysicalStandbymethod

• AnewSQLcommandwillarriveonthescene!

•  Encryptsalldatafilesinthetablespaceonline.– Primaryonly&Requiresauxiliarydiskspace

• CanalsobedoneOFFLINEforbetterperformance– NoExtrastorage,butrequiresdataoutageonPrimary.– OfflinecanbedoneonaPhysicalStandbyfirstaswiththepreviousexample

16

(>=12.2)

alter tablespace users encryption encrypt



• DGMGRLCommandImprovements• BrokerConfiguration-WideServiceName•  SimplifiedObserverManagementformultipleFast-StartFailoverconfigurations

17

BrokerGeneralEnhancements


NewDGMGRLcommands

•  EnhancedScriptingandExecution– Scriptexecutionvia‘@’– RepeatpreviousDGMGRLcommandwith‘/’– Executeoperatingsystemcommandwith‘HOST’or‘!’– Storeoutputinafilewith‘SPOOL’– Displaytimestampinpromptvia‘SETTIMEON/OFF’

18


Configuration-WideServiceName

• Providesasingleautomaticallystartedservicetoconnecttoanydatabase,PrimaryorStandby,inaDataGuardBrokerconfiguration.– DefaultstoPrimaryDBUniquenamewith‘_CFG’appended.– RequiredtoimplementtheSimplifiedObservermanagement(Nextslide)

• Canbesetbytheuserwith:

• Andseenwith:

19

edit configuration set property ConfigurationWideServiceName = ‘PROD_CFG’;

show configuration verbose ConfigurationWideServiceName; ConfigurationWideServiceName = ’PROD_CFG'


SimplifiedObserverManagement

•  Today(<12.2)whenyouhavemultipleBrokerconfigurationsusingFast-StartFailoveryoumustmanagetheObserversseparatelyeveniftheyallrunonthesame‘Observer’system.•  TheObserveralsorunsintheforeground•  Let’sfixthatlastonefirst

•  Thatsettlesthat• ButwhataboutmultipleFast-StartFailoverconfigurations?

20

start observer PRODObserver logfile is ‘/tmp/PRODObserver’ in background connect identifier is PROD;



• CreateanObserverConfigurationfile(herewehave4Observers,2groups)

•  TheCONNECT_IDmustpointtothePrimaryandallpotentialFSFOTargets

21

OneCommandtoRunandManageMultipleObservers

BROKER_CONFIGS = ( (CONFIG = (NAME=SALES) (CONNECT_ID=SALES_P)(FILES=/home)) (CONFIG = (NAME=HR) (CONNECT_ID=HR_P)(FILES=/home)) (CONFIG = (NAME=CUSTOMER) (CONNECT_ID=CUSTOMER_P)(FILES=/home)) (CONFIG = (NAME=ORDERS) (CONNECT_ID=ORDERS_P)(FILES=/home))) CONFIG_GROUPS = ( (GROUP =(NAME=GRP_A ) (CONFIG_LIST = (NAME=SALES)(NAME=ORDERS))) (GROUP= (NAME=GRP_B) (CONFIG_LIST = (NAME=HR)(NAME=CUSTOMER))))



•  SetthePropertytopointtoyourfileontheObserversystem

• Noticethatthereisnoneedtoconnecttoanydatabasetosetthisup

22


Oracle12c > dgmgrl DGMGRL for Linux: Release 12.2.0.0.0 Copyright (c) 1982, 2015, Oracle and/or its affiliates. All rights reserved. Welcome to DGMGRL, type "help" for information. DGMGRL> set observerconfigfile='/home/myobservers.ora'; DGMGRL> show observerconfigfile; ObserverConfigFile=/home/myobservers.ora



• OncethefileissetupinstructtheBrokertoruntheObservergroups

•  Therearenopasswordsintheconfigurationfile•  YoumustusetheWalletforauthenticationonalldatabasesinvolved.

23


DGMGRL> start observing grp_a ; ObserverConfigFile=/home/myobservers.ora Observer configuration file parsing succeeded Submitted command "START OBSERVER" using connect identifier “SALES_P" Submitted command "START OBSERVER" using connect identifier “HR_P“ Check superobserver.log and individual observer logs


ProtectionEnhancements12candBeyond

24


OracleDataGuard-Protection

•  EnhancedAutomaticBlockRecovery•  Zero-datalossFast-startFailoveruponstoragefailureeveninasyncmode– Ifinstancefails,shipsallredodataaccumulatedinlogbuffertostandby

• OracleDataGuardDatabaseCompare•  EnhancedAlternateDestinations• BrokerProtectionEnhancements• DisasterRecoveryintheCloud

25

EnhancedProtection


EnhancedAutomaticBlockRecovery

• ActiveDataGuardAutomaticBlockRepairintroducedin11gRelease2

• Blockslikefileheadersetcwerenotrepaired.•  InOracleDatabase12cRelease2thetypesofcorruptedblocksthatcanberepairedincreases!

26

ActiveStandbyDatabase

PrimaryDatabase

Read/WriteWorkload

Continuousredoshipping,validation&apply

Real-timeReporting


Zero-DataLoss(ZDL)Fast-StartFailover

•  Today,ZDLFast-StartFailoversareonlypossibleinSynchronousconfigurationsrunninginMaximumAvailability– Manytimesthough,failuresarestoragerelatedandnotcompletesystemfailures

•  InOracleDatabase12cRelease2,astoragefailureattheprimarydatabasewillbehandledbytheOracleDataGuardAsynchronousprocess.– Iftheinstancefails,itwillshipallredodataaccumulatedinlogbuffertostandby– ThiswillallowFast-StartFailovertoperformaZeroDataLossfailover•  EventhoughtheconfigurationwasinMaximumPerformance

– SimilartoanAutomaticIn-Memory‘ALTERDATABASEFLUSHREDO;’atfailuretime

27

Enhancedforasynchronousconfigurations


OracleDataGuardDatabaseCompare

• Highspeeddatabasecomparebetweenprimaryandstandby– DetectslatentcorruptionsonblocksthathavenotbeentouchedbyOracle•  Usefulindetectinglatentcorruptions,lostwrites•  Fastcompareatphysicallevel,accountsforblocksbeingmodified•  Network&computeefficiency– Send&compareblocksin“batches”– Optimizesforblockswhereversionsarethesame–  Ignoresemptyblocks

28

DGMGRL> validate database PROD1 datafile 10 output=checkPROD1; Operation requires a connection to database ”PROD1" Connecting ... Output files are created in /…/PROD1/trace on host “GRONAU1"


OracleDataGuardDatabaseCompareSummary: *************************************************************************** ID: Block Type Id TOTAL: Total number of blocks found DIFFV: Number of block pairs with different version LWLOC: Lost Writes at Local LWRMT: Lost Writes at Remote SAMEV: Number of block pairs with same version SAMEV&C: Number of block pairs with same version and checksum DIFFPAIR: Number of block pairs with same version but different contents ENCERR: Undecided blocks related to encryption/decrytion error. e.g. Wallet is not open. SKIPPED: Skipped blocks due to data corruption, etc ID TOTAL DIFFV LWLOC LWRMT SAMEV SAMEV&C DIFFPAIR ENCERR SKIPPED 29 0000001 0000000 0000000 0000000 0000001 0000001 0000000 0000000 0000000 30 0000125 0000000 0000000 0000000 0000125 0000125 0000000 0000000 0000000 58 0000512 0000512 0000000 0000000 0000000 0000000 0000000 0000000 0000000


EnhancedAlternateDestinations• Alternatedestinationsgotmuchbetterin12.1– Butfailbacktothefirstdestinationhappenedwhetherornotyouwantedit– Andyoulostfailbackifyouconfiguredmorethanonealternate.

•  Soin12.2wethrewawaytheALTERNATEattributeofredotransport– Replacingitwith2newattributes-GROUPandPRIORITY•  GROUPsaysweareallpotentialalternatesofeachotherinthesamegroup•  PRIORITYtellsDataGuardwhentofailbacktotheoriginaltargetbetweengroups

• Didyougetallthat?– Ididn’tthinkso,thisismygiveyouaheadachefeature.– Let’slookatanexample

30


ImprovedAlternateDestinations• Ourconfiguration– WehaveaPrimarywithtwoStandbys(TS1andTS2)•  Theycouldbeonelocalandoneremoteorbothremoteorbothlocal– ButwithFarSyncweshiponcefromthePrimaryandgetZDLatbothifnecessary

– Wealsohave2FarSyncinstancesthatcanservicethetwoStandbys(FS1,FS2)•  FarSync1&2areofequalconfigurationandbothlocal(withinSYNClatency)– SowecanloseoneandtheotherwilltakeoverandmaintainMaximumAvailability

•  ButifbothFarSyncInstancesaredownwestillwantredotobeshippedtoTS1&TS2•  Sowesetour4“LOG_ARCHIVE_DEST_n”parametersasfollows(allvalidforthePrimaryrole/logfile)

• Allclear?

31

2:SERVICE=FS1 SYNC GROUP=1 PRIORITY=1 3:SERVICE=FS2 SYNC GROUP=1 PRIORITY=1 4:SERVICE=TS1 ASYNC GROUP=1 PRIORITY=8 5:SERVICE=TS2 ASYNC GROUP=1 PRIORITY=8


ImprovedAlternateDestinations• OK,someEnglishtogowiththeparameters,theseareallthesamegroup

• ButthePrioritysays– SendRedotoFS1first,itforwardstheredotoTS1&TS2.– IfFS1failsstartsendingtoFS2whoforwardstheredotoTS1&TS2•  IfFS1comesbackdonothing,staywithFS2•  HoweverifFS2failsandFS1isback,reverttoFS1

– IfbothFS1andFS2failthensenddirectlytoTS1andTS2.•  SincetheyareconfiguredasASYNCtheProtectionLeveldropstoRESYNCHRONIZATION•  AfterthatifeitherFS1orFS2becomeavailablerevertbacktothem– ProtectionLevelgoesbacktoMaximumAvailability

32

2:SERVICE=FS1 SYNC GROUP=1 PRIORITY=1 3:SERVICE=FS2 SYNC GROUP=1 PRIORITY=1 4:SERVICE=TS1 ASYNC GROUP=1 PRIORITY=8 5:SERVICE=TS2 ASYNC GROUP=1 PRIORITY=8


ImprovedAlternateDestinations

• NormalSteadyState

33

Howaboutapicture?

PrimaryTS1FS1

TS2

FS2

SYNCASYNC



•  FarSyncFS1Fails(StayinPriority1)

34

Howaboutapicture?

PrimaryTS1FS1

TS2

FS2

SYNCASYNC



•  FarSyncFS1comesback(StayinPriority1,Nofailback)

35

Howaboutapicture?

PrimaryTS1FS1

TS2

FS2

SYNCASYNC



•  FarSyncFS2FailsandFS1isstillnotavailable(MovetoPriority8)

36

Howaboutapicture?

PrimaryTS1FS1

TS2

FS2

SYNCASYNC



•  FarSyncFS1comesback(MovebacktoPriority1)

37

Howaboutapicture?

PrimaryTS1FS1

TS2

FS2

SYNCASYNC


OracleDataGuard-Protection

• BlockComparisonToolSupportinDGMGRL– Shownintheearlierexample

• MultipleAutomaticFailoverTargets• MultipleObservers•  Fast-StartFailoverinMaximumProtectionMode•  SupportforEnhancedAlternateDestination•  SupportforTransportDestinationsofDifferentEndianess– Cross-endianDataGuard:AIXtoRecoveryAppliance

38

BrokerProtectionImprovements


PerformanceEnhancements12candBeyond

39


OracleDataGuard-Performance

• Multi-InstanceRedoApply– OracleDataGuardBrokerSupport• DatabaseIn-MemoryonanOracleActiveDataGuardStandbyDatabase• ActiveDataGuardsupportforDiagnosticPackfeatures• ActiveDataGuardSupportforSQLTuningAdvisor

40

PerformanceandTuningEnhancements


41

Single-InstanceRedoApply

MRPProcesses

MRPProcesses

StandbyInstance1

RFSProcess

CoordinatorProcess

MRPProcesses

StandbyInstance2

RFSProcess

CoordinatorProcess

MRPProcesses

StandbyInstance3

RFSProcess

PrimaryInstance1

ASYNC/SYNCProcess

PrimaryInstance2

ASYNC/SYNCProcess

PrimaryInstance3

ASYNC/SYNCProcess

Thread1Redo

Thread2Redo

Thread3Redo

SRL

SRL

SRL


42

Multi-InstanceRedoApply

CoordinatorProcess

MRPProcesses

StandbyInstance1

RFSProcess

CoordinatorProcess

MRPProcesses

StandbyInstance2

RFSProcess

CoordinatorProcess

MRPProcesses

StandbyInstance3

RFSProcess

PrimaryInstance1

ASYNC/SYNCProcess

PrimaryInstance2

ASYNC/SYNCProcess

PrimaryInstance3

ASYNC/SYNCProcess

Thread1Redo

Thread2Redo

Thread3Redo

SRL

SRL

SRL


Multi-InstanceRedoApply

• Parallel,multi-instancerecovery:standbywillkeepup– Standbyrecovery-utilizesCPUandIOacrossallnodesofRACstandby

– Haveseen3500MB+/secapplyrateonan8nodeRAC

• Multi-InstanceApplyrunsonallMOUNTEDinstancesorallOPENInstances•  ExposedintheBrokerwiththe‘ApplyInstances’propertyonastandbydatabase

43

recover managed standby database disconnect using instances 4;


DatabaseIn-MemoryandActiveDataGuard

•  In-memoryDBonActiveDataGuard– CreateIMCtablesandcolumnsforanalyticsonActiveDataGuard– Canpopulatewithdifferentdatathanproductiondatabase• Offloadevenmoretoyourstandby!

44

MonthIn-Memory

YearIn-Memory

StandbyPrimary


DiagnosticsandTuningforActiveDataGuard

•  SupportforDiagnosticPack(AWR)

•  SupportforTuningPackfeaturesandSQLPlanAnalyzer

45


AWRSupportforRemoteSnapshots

•  InOracleDatabase12cRelease2,theAWRframeworkisenhancedtosupportcaptureofremotesnapshotsfromanygenericdatabaseincludingActiveDataGuard(ADG)databases.– Atargetcatalogdatabasecollectssnapshotsfromtheremotedatabases(sources)– Snapshotscanbecollectedautomaticallyormanually– AWRtablesonthecatalogdatabaseaccumulatesnapshotdatafromallsourcesviadatabaselinks

– SourcedatabasesmustberegisteredonthecatalogvianewDBMS_WORKLOAD_REPOSITORY.REGISTER_REMOTE_DATABASEAPI

46

OnanActiveDataGuardStandby


SQLTuningAdvisorSupportforActiveDataGuard

• AllSQLTuningAdvisortasksissuedatthestandby– Createtuningtask,executetuningtaskandimplementSQLTuningAdvisorrecommendations

– Testexecution(heavylifting)happensonstandby•  Onlyminimalwriterelatedactivityonprimary

• Requireddatafetchedfromprimaryoveradatabaselinkfromstandby•  Taskdetailsandtuningresultsarestoredatprimaryandtheessentialdatarequiredtoconstructthereportisaccessedremotelyfromprimary•  Thereportisconstructedlocallyatthestandby,withnoCPUoverheadinprimary

47


Featurelist•  Fasterqueryresponsewithfasterredoapply– Multi-InstanceRedoApplyonADGnowsupportsIn-MemoryColumnar&Block-changetracking

•  ExtendADGenvironmentsforDataWarehousing– ADGsupportforNologgingworkloads•  FasterapplicationresponseafterADGroletransition– PreserveBufferCacheonADGafterroletransition• ProtectPrimarydatabasefromlostwritestoragefailures– NewLostwritedetectioncapabilityusingshadowtablespacesatPrimary

48


Multi-InstanceRedoApplyenhancements• Multi-InstanceRedoApplyallowsallstandbynodestoparticipateinrecovery•  In-memoryDB(IMC)onActiveDataGuardallows:*–  CreationofIMCtablesandcolumnsforanalyticsonActiveDataGuard–  Populationwithdifferentdatathanproductiondatabase–  Offloadingevenmoretoyourstandby!

• Multi-InstanceRedoapplyalsoworkswithBCT*AvailableonlyonExadataandOracleCloudOfferings

49

MonthIn-Memory

YearIn-Memory

StandbyPrimary


DataGuardandNoForceLogging*

•  ExtendedtoprovidebettersupportinanActiveDataGuardenvironmentwithoutsignificantlyincreasingtheamountofredogenerated.•  Twonewmodesareaddedasalternativestotheexistingnologgingmode– StandbyNologgingforLoadPerformance•  Ensuresthatstandbyswillreceivethenonloggeddatachangeswiththeminimumimpacttothespeedofloadingattheprimary– Thestandbycantransientlyhavenonloggedblocks.Thesenonloggedblockswillbeautomaticallyresolvedbymanagedstandbyrecovery.

– StandbyNologgingforDataavailability•  Ensuresallstandbyshavethedatawhentheprimaryloadcommitsbutatthecostofthrottlingthespeedofloadingdataattheprimary– Thestandbyswillneverhaveanynonloggedblocks.

50

*AvailableonEngineeredSystemsandOracleCloudonly


CreatingGTTsonActiveDataGuard• CreatingGTTstodaywillfailwhenexecutedonActiveDataGuard•  In18.1.0.0CREATE GLOBAL TEMPORARY TABLEonActiveDataGuardwill:– CreatetheGTTonthePrimary– WaitfortheGTTtobereplicatedandappliedtotheActiveDataGuardstandby– Returncontroltotheuser

51


ExecutingDMLonActiveDataGuard

• DMLRe-directionautomaticallyperformedfromActiveDataGuardstandbytothePrimarywithoutcompromisingACID– Enabledbysettingparameter“_enable_proxy_adg_redirect=TRUE”andappropriateconnectivityparameters•  StandbyADGsessionwaitsforupdatestoshowuponADGviaredoapply•  DMLdatare-directedbyanADGsessionisonlyvisibletothatsessionbeforecommit•  DMLdatare-directedisvisibletoallsessionsafterADGsessioncommits

– RespectsACIDpropertiesofadatabaseforADGsession– Targetedfor“ReadMostlyOccasionalUpdates”applications

52

RestrictedsupportforDMLRe-direction


NewBrokerDGMGRLcommands•  EnhancedScriptingandExecution– SETECHO|DEBUGON|OFF– SHOWALL•  ShowscurrentDGMGRLsettings

53


NewBrokerVALIDATEcommands• VALIDATEDATABASESPFILE– ValidateparameterssettingsbetweenthePrimaryandaStandbydatabase.

• VALIDATENETWORKCONFIG– Validatethenetworksetupbetweenalldatabasesintheconfiguration• VALIDATESTATICCONNECTIDENTIFIER– VerifythestaticconnectidentifiersetupinNon-CRSdatabases

54


Q&A


SafeHarborStatementTheprecedingisintendedtooutlineourgeneralproductdirection.Itisintendedforinformationpurposesonly,andmaynotbeincorporatedintoanycontract.Itisnotacommitmenttodeliveranymaterial,code,orfunctionality,andshouldnotberelieduponinmakingpurchasingdecisions.Thedevelopment,release,andtimingofanyfeaturesorfunctionalitydescribedforOracle’sproductsremainsatthesolediscretionofOracle.

56


Oracle Data Guard 12.2 - 18c - doag.org · • In Oracle Database 12c Release 2, a storage failure at the primary database will be handled by the Oracle Data Guard Asynchronous process.

Documents