Sqlserver Ha

8/7/2019 Sqlserver Ha

1/32

SQL ServerHigh Availability

Craig RyanNational Manager Database Services


2/32


3/32

How MuchAvailability do you need?

Need to ask yourself:

How long can we afford to be down?

How much data can we afford to lose?

Availability = up/(up+down)

Category % Uptime

Downtime

(Min/mth)

Downtime

(Hrs/mth)

Downtime

(Min/Year)

Downtime

(Hrs/Year)

1 nine 98.000% 876.00 14.600 10,512 175.20

2 nines 99.000% 438.00 7.300 5,256 87.60

3 nines 99.900% 43.80 0.730 526 8.76

4 nines 99.990% 4.38 0.073 53 0.88

5 nines 99.999% 0.44 0.007 5 0.09


4/32

Barriers ToAvailability

Database Server Failure or Disaster User or Application Error

Data Access Concurrency Limitations

Database Maintenance and Operations

Upgrades

Availability at Scale

Many barriersMany barriersOnly some are addressable by DBMS technologyOnly some are addressable by DBMS technology

Be sure to consider people, planning, and proceduresBe sure to consider people, planning, and procedures


5/32

Warm Standby SolutionsReplication and Log Shipping

Database Object Level

SQL Server Replication

Database Level

Log Shipping

Both provide multiple copies

and a MANUAL fail over


6/32

Log Shipping Minimal impact on the production

server

No changes to the database arerequired

Transactional consistency

Supports delayed load oftransaction logs

Not all SQL Server objects areautomatically copied

Users must exit for next log to beapplied

Log ShippingLog Shipping

Primary

SQLServer

Secondary

SQL Server

Tra

Log

Tra

Log

Load Job

opy Job


7/32

DemoSQL 2000 Log Shipping


8/32

Replication Failover possible; a custom

solution Not limited to entire database; Can

define subset of source database

or tables

Copy of database is continuouslyaccessible for read activity

Latency between source and copy

can be as low as seconds

Significantly increases the on-

going Database management

Database schema may need to be

changed

ReplicationReplication


9/32

DemoSQL 2000 Transactional Replication


10/32

Hot Standby Automatic failover

Built on Microsoft Cluster Services (MSCS)

Multiple nodes provide availability, transparent to client

Automatic detection and failover

Requires certified hardware

Supports many scenarios: Active/Active, N+1, N+I

Instance Failover entire instance works as a unit

Single copy of instance databases Available since SQL Server 7.0

Standby is not available for reporting, queries, etc.

May support other instances

Failover ClusteringMicrosoft Cluster Services

Failover ClusterFailover Cluster


11/32

Client CsClient Cs

Server AServer A Server BServer B

SharedShared

Disk ArrayDisk Array

HeartbeatHeartbeatCluster managementCluster management

SQL ServerSQL Server

Virtual

Server

EE FF GGC,DC,D C,DC,D

SQL Server

HubHub HubHub

Active/Passive SQL Server Cluster


12/32

Client CsClient Cs

Server AServer A Server BServer B

SharedShared

Disk ArrayDisk Array

HeartbeatHeartbeatCluster managementCluster management

SQL ServerSQL Server

Virtual

Server

EE FF GGC,DC,D C,DC,D

HubHub HubHub

SQL Server

Active/Passive SQL Server Cluster


13/32

What Clustering doesnt do:

Clustering is not a mechanism to scale

Doesnt protect your server against siteoutage

Doesnt protect your disk subsystem

Doesnt protect against databasecorruption

Doesnt protect against logical corruption

Doesnt protect against user error

Doesnt protect application crash

Clustering is not a method to load-balance

Still a single point of failure The Database!


14/32

DemoSQL 2000 Failover Clustering


15/32

Log Shipping vs Clustering vs SQL Replication

So whats the best solution.

On your business requirements

it depends

You can combine the SQL H.A options.

E.g. A/ Cluster with Log shipping


16/32

SQL Server 2005High Availability


17/32

Barriers ToAvailabilityAs addressed in SQL Server 2005As addressed in SQL Server 2005

Database Server Failure or Disaster

Failover Clustering

Database Mirroring Transparent Client Redirect

User or Application Error

Data Access Concurrency Limitations

Database Maintenance andOperations

Availability at Scale


18/32

Failover ClusteringSQL Server 2005SQL Server 2005

Further refined in SQL Server 2005

More nodes

Match operating system limits

Unattended setup

Support for mounted volumes (Mount oints)

All SQL Server services participate Database Engine, SQL Server Agent,

Analysis Services, Full-Text Search, etc.

Failover ClusterFailover Cluster


19/32

Database MirroringNewfor SQL Server 2005

Database MirroringNewfor SQL Server 2005

Instant Standby

Conceptually a fault-tolerant server

Database Failover Very Fast less than three seconds

Zero data loss

Automatic or manual failover

Automatic re-sync after failover

Automatic, transparent client redirect

Database MirroringDatabase Mirroring


20/32

Database Mirroring

Hardware

Works with standard computers,

storage, and networks

No shared storage components,

virtually no distance limitations

Impact to transaction throughput

Zero to minimal, depending on

environment / workload

Database MirroringDatabase Mirroring


21/32

Database MirroringHow itworks

MirrorMirrorrincipalrincipal

WitnessWitness

Log

Application

SQL Server SQL Server

2

2

4

51

Data DataLog

3>2 >3

Mirror is alwaysMirror is always

redoingredoing itit

remains currentremains currentCommit


22/32

Witness and Quorum

Sole purpose of the Witness isto provide automatic failover

To survive the loss ofone server you

must have at least three Prevents split brain

Does a lost connection mean the partner is

down or is the network down?

To become the Principal, a server must

talk to at least one other server

WitnessWitness


23/32

Witness

Witness is an instance of

SQL Server 2005

Single witness for multiple sessions

Consumes very little resources

Not a single point of failure

Partners can form quorum on their own

WitnessWitness


24/32

Safety / Performance

There is a trade-off betweenperformance and safety

Database Mirroring has two safety

levels FULL commit when logged on Mirror Allows automatic failover

No data loss

OFF commit when logged on Principal System does its best to keep up

Prevents failover; to make mirror available

Must force service

Or terminate Database Mirroring session


25/32

Transparent Client Redirect

No changes to application code Client automatically redirected if

session is dropped

Client library is aware ofPrincipal andMirror servers

Upon initial connect to Principal, library

caches Mirror name

When client attempts to reconnect

IfPrincipal is available, connects

If not, client library automatically redirects

connection to Mirror


26/32

Database Mirroring Setup Steps1. Ensure SQL 2005 is installed on both the

principle & the mirror Server2. Setup Security (Endpoints)

3. Prepare the mirror database

4. Setup & Start the database mirroring session

5. Optionally add a witness


27/32

DemoSQL 2005 Database Mirroring


28/32

Database States for Database Mirroring SYNCHRONIZING

SYNCHRONIZED SUSPENDED

PENDING_FAILOVER

DISCONNECTED


29/32

Automatic Failover Automatic Failover requires the following

conditions: Database mirroring running in synchronous mode

The database must be in a synchronised state

A witness must exist

During Failover the following actions occur:

The witness & mirror server agree the primary is dead,which puts the database into a suspend state

If possible the database on the Primary server changesto a disconnected state

The mirror finishes rolling forward and records the LSN

The mirror database comes online

When the principle returns, it becomes the mirror


30/32

Failover SolutionsAtAGlance

Clustering & Mirroring both provide:

Automatic detection and failover

Manual failover

Transparent client connect

Database Mirroring

Database scope

Standard servers

Fastest failover

Limited reporting on standby

Duplicate copy of database

Failover Clustering

System scope

Certified hardware

Fast failover

No reporting on standby

Single copy of database


31/32

Summary

How many 9s do you NEED?

SQL 2000 potential downtime ~3 mins

SQL 2005 potential downtime ~3 secs

Remember to consider process too!


32/32

Feel free to contact us

Craig RyanNational Manager Database Services

[email protected]

Andrew Gannon

Business Development [email protected]

(03) 9427-1477

Documents

Sqlserver Ha