在 Oracle 19c 标准版 2 中实现高可用性和灾难恢复的更好方法.pdf

上善若水

6页

2次

2022-06-12

免费下载

Has Oracle 19c Standard Edition 2 (SE2) taken a step backward

in high availability (HA) and disaster recovery (DR)?

It appears so. If you plan to stick with Standard Edition 2, take

a second look at how you will handle server failures and avoid

downtime in your environment. Oracle 19c will make you re-think

your options for achieving the high availability and disaster

recovery you’ve become accustomed to.

This technical brief examines those options in detail. You’ll see

the dierences — especially in recovery time — and discover

how you can use SharePlex® by Quest® to stick with Oracle SE2

without putting your high availability and disaster recovery

strategies at risk.

“WAIT — WHAT HAPPENED TO SUPPORT FOR MY RAC?”

If you are currently using Oracle SE2 or considering it to reduce

your Oracle spend, you probably know that its high availability

and disaster recovery options are limited. In fact, since Data

Guard is not available for SE, you have to look for disaster

recovery tools from other sources. Oracle Real Application

Clusters (RAC) is your only native option for high availability.

Or at least it has been. RAC is a de-supported feature for the

Standard Edition 2 of Oracle 19c. Your availability options

in Oracle 19c SE2 have been reduced to Oracle Fail Safe

(Windows only) and a new oering called Standard Edition High

Availability (SE HA).

On close inspection, though, Oracle SE HA reveals itself to be

a downgrade from RAC:

“

Standard Edition High Availability databases are not Real

Application Clusters (RAC)-enabled. Oracle RAC One Node is a

RAC-enabled database in that the RAC-option needs to be enabled

in the Oracle Database home from which the database runs. This is

not the case for Standard Edition High Availability databases.”

So that’s what happened to support for your RAC in Oracle

Standard Edition 2. If you want it back, you’ll have to upgrade to

Oracle 19c Enterprise Edition.

A Better Way for High

Availability and Disaster

Recovery in Oracle 19c

Standard Edition 2

What if you could implement reliable high availability and disaster recovery in

Oracle 19c SE2?

By Clay Jackson, Database Systems Engineer, Quest®

The documentation states, “Upgrading

Oracle Database Standard Edition data-

bases that use Oracle Real Application

Clusters (Oracle RAC) functionality from

earlier releases to Oracle Database 19c

is not possible. To upgrade those data-

bases to Oracle Database 19c, either

remove the Oracle RAC functionality

before starting the upgrade, or upgrade

from Oracle Database Standard Edition

to Oracle Database Enterprise Edition.”

ORACLE 19C SE HA AND ORACLE

RAC FOR SE ARE COMPARABLE . . .

But let’s say that upgrade is not in your

budget. That means you’ll be left with the

options in Oracle 19c SE HA.

Oracle RAC for SE delivered high availability

within a single data center with two running

nodes supporting a single database. Oracle

19c SE HA is a set of packaged scripts that

run on top of Oracle Clusterware and use

Oracle Grid Infrastructure to manage failover.

And, if you look closer, you’ll see that SE

HA has the same data center limitations

and the same points of failure as RAC.

SE HA uses Oracle Clusterware and Oracle

Automatic Storage Management (ASM). That

means that both the primary and standby

(or failover) servers have to be located in

the same data center and connected to the

same shared storage. Then, just like RAC,

SE HA uses the same database. If the ﬁles

containing your data are unavailable, then

your database will be unavailable.

So, in that regard, SE HA and RAC are

comparable.

. . . BUT THE FAILOVER IN SE HA

IS THE BIG STEP BACKWARD

When you consider recovery times,

however, it’s clear that SE HA is a down

grade from RAC.

RAC is a clustered database with a shared

cache architecture, so all servers in the

cluster pool are active and continuously

processing trac. If a server in the clus

ter pool fails, the database continues to

process activity on the remaining nodes.

In other words, there is no actual failover

with RAC — you get continuous availabil

ity when handling server failures.

On the other hand, if you do choose to

adopt Oracle 19c SE HA and a server fails,

count on the following worst-case scenario:

• The instance will crash.

• Oracle Grid Infrastructure will need time

to identify the crash. Then, it will mount

the database disks to a designated server.

Finally, Oracle Grid Infrastructure will

start the instance, triggering a recovery.

• The instance will become available for users.

RAC doesn’t introduce failover delays;

SE HA does. It’s important to test the

failover in SE HA because this type of

failure recovery could take a long time

in your environment. In any event, it will

certainly take longer than it would have

taken with RAC.

When you consider

recovery times, it’s

clear that SE HA is

a downgrade from

RAC. RAC doesn’t

introduce failover

delays; SE HA does.

SE SE