Scribd
Upload
132 views4 pages

28 2 Always On Clusters Notes

This document discusses the steps to configure an Always On Availability Group (AOAG) for database high availability and disaster recovery. Key steps include creating an AOAG in a SQL Server failover cluster, adding databases and secondary replicas, configuring automated backup preferences and data synchronization, and adding an AOAG listener and registering it with DNS to direct connections to the AOAG virtual IP address.

Uploaded by

Hari Prasad Mullamuri
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
132 views4 pages

28 2 Always On Clusters Notes

This document discusses the steps to configure an Always On Availability Group (AOAG) for database high availability and disaster recovery. Key steps include creating an AOAG in a SQL Server failover cluster, adding databases and secondary replicas, configuring automated backup preferences and data synchronization, and adding an AOAG listener and registering it with DNS to direct connections to the AOAG virtual IP address.

Uploaded by

Hari Prasad Mullamuri
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
You are on page 1/ 4

ALWAYS-ON AVAILABILTY GROUPS [AOAG]:

THIS IS A HA-DR MECHANISM, APPLICABLE FOR MULTI DATABASE REPLICATIONS


TO BE CONFIGURED IN A CLUSTERED ENVIRONMENT ONLY.
IN ALWAYS-ON WE REPLICATE THE ENTIRE DATABASE RATHER THAN INDIVIDUAL OBJECTS.
MEANS : DATABASE LEVEL ENCRYPTED OBJECTS AND DATABASE LEVEL USERS CAN ALSO BE
REPLICATED.
FURTHER, WE DO NOT USE SQL SERVER AGENT JOBS. INSTEAD, WE USE "ENDPOINTS"
END POINTS ARE COMPUTER PORTS USED FOR DIRECT DATA COMMUNICATION TO THE
APPLICATION.
USING ALWAYS-ON WE CAN ENSURE MINIMAL OR NO DATA LOSS WITH AUTOMATED FAILOVER
SUPPORT.

PRACTICE STEPS [OVERVIEW]:


STEP 1: LAUNCH VIRTUAL LAB > CONNECT TO DC MACHINE
STEP 2: LAUNCH SSMS TOOL FROM DC MACHINE
STEP 3: CONNECT TO EVERY CLUSTER INSTANCE [SQL SERVER INSTANCE]
USING SSMS TOOL FROM DC MACHINE
STEP 4: FROM 1ST SERVER: CREATE TWO NEW DATABASES.
BACKUP ALL DATABASES ON THE SERVER.
RIGHT CLICK ALWAYS-ON AVAILABILITY GROUP FOLDER > SPECIFY NAME TO
AG GROUP > SELECT REQUIRED DATABASES >
SPECIFY REPLICAS [ADD SUBSCRIBERS, AUTOMATED FAILOVER,
SYNCHRONOUS DATA TRANSFER, READABLE SECONDARY]
DEFINE BACKUP LOCATION AND INITIALIZATION [AUTOMATED SEEDING] >
FINISH.
STEP 5: LAUNCH ALWAYS-ON DASHBOARD > MONITOR
RIGHT CLICK ALWAYS-ON > ADD DATABASE > RUN THRU THE WIZARD TO ADD
NEW DATABASE TO EXISTING LIVE AOAG CONFIGURATION

STEP 6: RIGHT CLICK ALWAYS-ON > ADD LISTNER > SPECIFY STATIC IP
ADDRESS.
LAUNCH SERVER MANAGER > TOOLS > DNS > FORWARD LOOKUP ZONE >
ACTION : ADD NEW HOST [AA] > SPECIFY ABOVE IP ADDRESS.

PRACTICE STEPS [DETAILED]:

STEP 1: VISIT THIS VIRTUAL LAB LINK:


https://labondemand.com/Event/MsHandsOnLabs/Home

SEARCH CODE: SQ00145

WE SEE TOTAL OF 4 COMPUTERS IN THIS VIRTUAL ENVIRONMENT.


COMPUTER 1: DOMAIN CONTROLLER. INSTALLED WITH ACTIVE
DIRECTORY [AD], SQL SERVER MANAGEMENT STUDIO TOOL

COMPUTER 2,3,4: CLUSTER NODES INSTALLED WITH SQL SERVER


2016 DEFAULT INSTANCE. NO SSMS TOOL INSTALLED.

STEP 2: LOG IN TO THE DOMAIN CONTROLLER MACHINE [DC MACHINE] WITH :


LAB USER
Pass@word1

STEP 3: FROM THE DC MACHINE : LAUNCH SSMS TOOL > CONNECT TO BELOW THREE
SQL SERVERS:
WIN-AGNODE01
WIN-AGNODE02
WIN-AGNODE03

STEP 4: ON NODE 1 SERVER [WIN-AGNODE01] : WE SEE ONE DATABASE "Expenses".


CREATE TWO MORE NEW DATABASES :
DATABASE2
DATABASE3

STEP 5: ON NODE 1 SERVER [WIN-AGNODE01] : PERFORM BACKUP OF ALL THREE DATABASES


TO THE DEFAULT PATH
NOTE THAT NODE 2 AND NODE 3 DO NOT HAVE ANY USER DATABASES.

STEP 6: FROM NODE 1 SERVER [WIN-AGNODE01] > ALWAYS ON HIGH AVAILABILITY >
RIGHT CLICK > NEW AVAIALBILITY GROUP WIZARD >
SPECIFY THE GROUP NAME, SELECT WINDOWS SERVER FAILOVER CLUSTER > CHECK
@ HEALTHCHECK OPTION AND DTC OPTION
SELECT Expenses Database & DATABASE1 Database > CURRENT SERVER [NODE1]
IS AUTO LISTED AS PRIMARY.
CLICK @ ADD REPLICA > CONNECT TO NODE2 > THEN NODE2 SERVER IS AUTO
LISTED AS SECONDARY.
SELECT AUTOMATIC FAILOVER FOR NODE 1 & NODE 2
AVAILABILITY MODE : SYNCHRONOUS COMMIT
READABLE SECONDARY : No : No user connections are allowed to
secondary databases of this replica.

Read-intent only : APPLICABLE FOR RPC


Only read-only connections are allowed to
secondary databases of this replica.

Yes : APPLICABLE FOR RPC OUT


All connections are allowed to secondary
databases of this replica, but only for read
access. The secondary database(s) are all available for read access.

NOTE: IN AAG 2014 AND 2012 : WE HAVE OPTION FOR


"DUAL WRITE" MECHANISM.
IN THESE VERIONS, SECONDARY SERVERS CAN ALSO WRITE
DATA BACK TO PRIMARY IF NEEDED.

END POINTS : LEAVE THEM TO DEFAULT [5022 END POINT ON EACH ALWAYS ON
NODE]
END POINT NAME : HADR_ENDPOINT [USED FOR ENCRYPTED DATA
TRANSFER BETWEEN THE REPLICAS]

SPECIFY BACKUP PREFERENCES:


** 1. PREFER SECONDARY : AUTOMATED AVAILABILITY BACKUPS
OCCUR ON SECONDARY. IF NO SECONDARY IS ONLINE THEN
THE BACKUPS OCCUR ON PRIMARY.

2. SECONDARY ONLY : AUTOMATED AVAILABILITY BACKUPS OCCUR ON


SECONDARY. IF NO SECONDARY IS ONLINE THEN
NO BACKUPS OCCUR.

3. PRIMARY : AUTOMATED AVAILABILITY BACKUPS OCCUR ON


PRIMARY.

4. ANY REPLICA : AUTOMATED AVAILABILITY BACKUPS


OCCUR ON PRIMARY. IF NOT ON SECONDARY.

CLICK @ NEXT.
SELECT INITITAL DATA SYNCHRONIZATION OPTION:
** 1. AUTOMATIC SEEDING : SECONDARY
DATABASES ARE AUTO CREATED.
2. FULL DATABASE BACKUP & LOG BACKUP : SECONDARY
DATABAES ARE RESTORED USING PRIMARY BACKUPS
3. JOIN ONLY : NO SECONDARY
DATABASES ARE CREATED.
4. SKIPP INITIAL DATA SYNCHRONIZATION : ALWAYS ON IS
NOT CONFIGURED COMPLETELY. BUT SECONDARY
ENTRIES ARE ADDED.

CLICK @ NEXT. IGNORE THE LISTENER CONFIGURATION WARNING.

STEP 7: HOW TO VERIFY ALWAYS - ON CONFIGURATION?


HOW TO MONITOR AOAG CONFIGURATION?

FROM THE PRIMARY NODE > RIGHT CLICK ALWAYS ON AVAILABILTY GROUPS FOLDER
> SHOW DASHBAORD.
THEN FROM THE DASHBAORD > CLICK @ AVAILABILOITY GROUP > THIS GIVES THE
DETAILS OF THE AVAIALBILITY GROUP:
1. AVAILABILITY GROUP STATE
2. PRIMARY INSTANCE
3. SECONDARY INSTANCE
4. FAILOVER MODE
5. CLUSTER STATE
6. CLUSTER TYPE
7. AVAILABILITY REPLICAS
8. DATABASES ON PRIMARY & SECONDARY
NAME, REPLCA, SYNC STATE, FAILOVER REDINESS ISSUES

9. ALWAYS-ON HEALTHCHECK EVENTS

STEP 8: HOW TO ADD A NEW SECONDARY TO EXISTING (LIVE) FAILOVER CLUSTER ?


FROM PRIMARY > EXPAND ALWAYS ON AVAILABILITY GROUPS FOLDER > GO TO
AVAIABILITY GROUP CREATED ABOVE >
RIGHT CLICK > ADD REPLICA > CONNEC TO EXISTING SECONDARY REPLICA(S) >
ADD REPLICA > SPECIFY 3RD NODE >
SYNCHRONOUS COMMIT, AUTOMATIC FAILOVER, READABLE SECONDARY : YES > DATA
SYNCHRONIZATION > FINISH

STEP 9: HOW TO ADD A NEW DATABASE TO EXISTING (LIVE) ALWAYS ON AVAILABILITY


GROUP ?
FROM PRIMARY > EXPAND ALWAYS ON AVAILABILITY GROUPS FOLDER > GO TO
AVAIABILITY GROUP CREATED ABOVE >
RIGHT CLICK > ADD DATABASE > CONNEC TO EXISTING SECONDARY REPLICA(S) >
SELECT THE DATABASE 3 > DATA SYNCHRONIZATION >
FINISH.
STEP 10: HOW TO TEST THE FAILOVER?
FROM ALWAYS ON DASHBAORD SCREEN > START FAILVOER WIZARD > SELECT
REQUIRED SECONDARY THAT SHOULD BE THE NEW PRIAMRY >
CONNECT TO THE SECONDARY SERVER > FAILOVER.

STEP 11: HOW TO ADD ALWAYS ON AVAILABILITY GROUP [AG] LISTENER AND LINK TO DNS ?
LISTENERS ARE THE GATEWAYS TO CONNECT THE CURRENT ALWAYS-ON SQL SERVER
CLUSTER WITH THE OS LEVEL DOMAIN IN DC MACHINE.
THIS CONNECTIONS NEEDS TO BE REGISTERED WITH "DOMAIN NAME SERVER : DNS"
AT THE DOMAIN CONTROLLER LEVEL.

FROM PRIMARY > EXPAND ALWAYS ON AVAILABILITY GROUPS FOLDER > GO TO


AVAIABILITY GROUP CREATED ABOVE >
RIGHT CLICK > ADD LISTENER > SPECIFY A NAME.
PORT : 1433
STATIC IP ADDRESS
SUBNET : 10.0.2.0/24
IPV4 : 10.0.2.125

GO TO SERVER MANAGER > TOOLS > DNS > FORWARD LOOKUP ZONES >
AOAG.SQLSERVER.LABS > ACTION > NEW HOST > SPECIFY ABOVE IP.
THEN WE SEE ONE FORMAT FOR ABOVE DEFINED AOAG ENVIRONMENT:
<domainname>.aoag.sqlserver.labs ip#: <
> = VIP = Virtual Ip Address

You might also like