 | Level: Intermediate Hidayatullah Shaikh (hshaikh@us.ibm.com), Senior Software Engineer, IBM Watson Research Center
14 Jul 2005 Data is at the heart of any business, and access to it should be available with minimum downtime. In this article, take a look at the setup and implementation of a Linux™ High Availability solution for IBM DB2® Universal Database® -- the database management system that delivers a flexible and cost-effective database platform for building robust, on demand business applications. Using this step-by-step guide, you can set up and run a highly available DB2 UDB database.
Data is the blood of a modern, on demand business; the system that stores and moves that data around -- servers, networks, databases -- is the heart. But both are inert components without the heartbeat -- having reliable, readily available access to that data, access with minimum downtime.
Introduction
The first installment of this series, High-availability middleware on Linux, Part 1: Heartbeat and Apache Web server, introduced high availability (HA) concepts and how to install and configure heartbeat. This part, the final installment of the series, covers the HA implementation for DB2 UDB 8.1 in a cold standby configuration using heartbeat.
|
Listen for the heartbeat
Heartbeat is one of the publicly available packages of the Linux-HA project (for a link, see the Resources section later in this article). Heartbeat provides the basic functions required by any HA system, such as starting and stopping resources, monitoring the availability of the systems in the cluster, and transferring ownership of a shared IP address between nodes in the cluster. Heartbeat also monitors the health of a particular service (or services) through a serial line or Ethernet interface or both. The current version supports a two-node configuration where special heartbeat "pings" are used to check the status and availability of a service.
|
|
In this implementation, heartbeat detects a failure with the primary and initiates failover by:
-
Stopping the DB2 processes on the primary
-
Releasing the shared disk on the primary
-
Removing the service IP address on the primary
-
Adding the service IP address to the standby
-
Mounting the shared disk on the standby
-
Restarting the DB2 processes on the standby machine
To get the most from this article, you need a basic understanding of DB2 UDB and high availability clusters. The first article in this series explains what it means for software to be highly available and how to install and set up heartbeat software from the High-Availability Linux project on a two-node system.
DB2 UDB and HA basics
Any DB2 UDB that is to be used in a heartbeat cluster needs to have its data on shared disks so that it can be accessed by a surviving node in the event of a node failure. A node running a database instance must also maintain a number of files on internal disks. These files include those that relate to all the database instances on the node.
|
High Availability Disaster Recovery (HADR)
High Availability Disaster Recovery (HADR) is a new feature in DB2 UDB V8.2 that provides a high-availability solution for protection against both partial and complete site failures. Using the HADR setup wizard, you can set up the entire configuration in a matter of minutes. HADR provides high availability for your databases by replicating data from a source (primary) database to a target (standby) database and keeping the two databases in sync. When a failure occurs at the primary database server, the standby database can take over and become the primary database in seconds with just a simple DB2 command or a click of a button in the HADR management wizard.
Combined with the new automatic client reroute capability, HADR provides transparency to the application regardless of the failure type, whether it is hardware, network, software issues, or disaster scenarios such as fire.
Using cluster management software such as Tivoli System automation for Multiplatforms included with DB2 UDB on Linux, the failover step can be completely automated, eliminating the need for DBA intervention. To see how this is done, read the technical paper "Automating DB2 HADR Failover on Linux using Tivoli System Automation" (PDF). Alternatively, the open source Heartbeat Linux-HA package can be used to provide the monitoring and automatic failover capabilities.
Learn more about HADR on the IBM DB2 Web site.
|
|
Files related to a database instance are therefore divided between internal and shared disks. Figure 1 details this organization of filesystem for DB2, described in our test setup for instance db2inst1 and database hadb.
Figure 1. Setup for high availability DB2 for instance db2inst1 and database hadb
In this setup:
-
The machine ha1 will serve as a primary DB2 UDB database machine.
-
The machine ha2 will serve as a backup for node ha1.
-
Each node will have its own local copy of DB2 UDB 8.1 installed.
-
The database hadb-specific directories (db2inst1/NODE0000/SQL00001 and db2inst1/NODE0000/sqldbdir) will be kept on the shared filesystem (/ha).
Installing the database
Follow the steps in this section to install DB2 UDB 8.1 on both the primary and the backup node. For more information, refer to the DB2 Information Center (see Resources):
-
Log in as root.
-
Extract the DB2 UDB 8.1 installation image using these commands:
rm -rf /tmp/db28.1-install
mkdir /tmp/db28.1-install
tar xf C48THML.tar -C /tmp/db28.1-install
|
Here C48THML.tar is the installation tar file.
-
Set the kernel level:
export LD_ASSUME_KERNEL=2.4.19.
-
Do not use the IBM Developer Kit for Linux, Java 2 Technology Edition provided in the DB2 Installation disk. Replace the JDK that comes with DB2 with the IBM 1.4.2 JDK.
cd /tmp/db28.1-install/009_ESE_LNX_32_NLV
mv ./db2/linux/java ./db2/linux/java.db2
ln -s /opt/IBMJava2-142 ./db2/linux/java
|
-
Launch the DB2 setup wizard using this command:
./db2setup.
Use the following information in the wizard screens:
-
For Product to install use DB2 UDB Enterprise Server Edition.
-
For Group and User IDs, the values of the numerical group ID (gid) and the user ID (uid) fields must match on both machines. We will use the IDs show in Table 1 below.
-
For Partition, select single-partition instance.
-
For DB2 Instance Name, select db2inst1.
Table 1. Group and user names and IDs for DB2 setup
|
Group name
|
GID
|
User name
|
UID
| | dasadm1 | 2001 | dasusr1 | 2001 | | db2grp1 | 2002 | db2inst1 | 2002 | | db2fgrp1 | 2003 | db2fenc1 | 2003 |
Creating a highly available database
Follow these steps to create the highly available database, hadb:
-
Log on as user db2inst1 on both the primary (ha1) and backup (ha2) nodes:
su - db2inst1.
-
Ensure that DB2(R) instances are not started at boot time by using the
db2iauto utility as follows (as user db2inst1) on both nodes ha1 and ha2:
cd sqllib/bin
./db2iauto -off db2inst1
|
-
Modify the /etc/inittab file to run DB2 HA correctly. Comment out the line that starts DB2 on system startup as follows on both nodes ha1 and ha:
#fmc:2345:respawn:/opt/IBM/db2/V8.1/bin/db2fmcd #DB2 Fault Monitor Coordinator.
-
Start DB2 on the primary node ha1:
db2start.
-
Mount the filesystem /ha on the backup (ha1) using this command (as user root):
mount /ha.
-
Create a database hadb using this command on node ha1:
db2 create database hadb on /ha.
-
Make sure you are able to connect to the database hadb using this command on node ha1:
db2 connect to hadb. If successful, disconnect using this command: db2 connect reset.
-
Stop DB2 using
db2stop on node ha1.
-
Mount the filesystem /ha on the backup (ha2) using
mount /ha as user root.
-
Start DB2 on the primary node ha2 with
db2start.
-
As user db2inst1, execute the following command on node ha2 in order to catalog the database hadb:
db2 catalog database hadb on /ha.
-
Make sure you are able to connect to the database hadb by using
db2 connect to hadb on node ha2. If successful, disconnect using db2 connect reset.
-
Stop DB2 using the
db2stop on node ha2.
 |
Configuring heartbeat to manage DB2
Now configure the /etc/ha.d/haresources file (on both the primary and backup machine nodes) to include the script that manages the DB2 processes. This script comes with heartbeat.
The relevant portion of the modified file is shown here:
ha1.haw2.ibm.com 9.22.7.46
Filesystem::hanfs.haw2.ibm.com:/ha::/ha::nfs::rw,hard db2::db2inst1
|
This line dictates that on startup of heartbeat, ha1 serves the cluster IP address, mounts the shared filesystem, and starts the database server as well. On shutdown, heartbeat first stops the database server, then un-mounts the shared filesystem, and finally gives up the IP.
Testing DB2 UDB failover
This section shows you how to test the high availability of the DB2 database, hadb. This is perhaps the most involved process described in this article, so watch the details.
-
Start the heartbeat service on the primary and then on the backup node. Use the command /etc/rc.d/init.d/heartbeat start as user root.
After heartbeat starts successfully, you should see a new interface with the IP address that you configured in the ha.cf file. Once you've started heartbeat, take a peek at your log file (default is /var/log/ha-log) on the primary, and make sure that it is doing the IP takeover and then starting DB2. Use the ps command to make sure DB2 processes are running on the primary node. Heartbeat will not start any of the above processes on the backup. This happens only after the primary fails.
-
Check the database status on ha1 node as user db2inst1:
db2 -tf /ha/hahbcode/db2/listdb.sql. The output of this command for our run should be the following:
Active Databases
Database name = HADB
Applications connected currently = 0
Database path = /ha/db2inst1/NODE0000/SQL00001/
|
-
Create a test table (hadb.TestHATable) on ha1 node as user db2inst1, like so:
db2 -tf /ha/hahbcode/db2/createdb.sql.
-
Now, insert a row of data in the test table on ha1 node as user db2inst1:
db2 -tf /ha/hahbcode/db2/insertdb.sql.
-
Inspect the contents of the test table on ha1 node as user db2inst1:
db2 -tf /ha/hahbcode/db2/selectdb.sql. You should be able to see the row of data that you inserted in the previous step. The output for our run is:
[db2inst1@ha1 db2inst1]$ db2 -tf /ha/hahbcode/db2/selectdb.sql
Database Connection Information
Database server = DB2/LINUX 8.1.0
SQL authorization ID = DB2INST1
Local database alias = HADB
COL1 COL2
----------- ------------
10 Hello
1 record(s) selected.
DB20000I The SQL command completed successfully.
|
-
To simulate failover, we simply stop heartbeat on the primary system as user root:
/etc/rc.d/init.d/heartbeat stop. You should see all the services come up on the backup machine in under a minute. You can verify that DB2 is running on the backup by checking the /var/log/ha-log file and using the ps command on the backup machine.
-
Check the database status on ha2 node as user db2inst1:
db2 -tf /ha/hahbcode/db2/listdb.sql. The output of this command for our run is:
Active Databases
Database name = HADB
Applications connected currently = 0
Database path = /ha/db2inst1/NODE0000/SQL00001/
|
-
Inspect the contents of the test table (hadb.TestHATable) on ha2 node as user db2inst1:
db2 -tf /ha/hahbcode/db2/selectdb.sql. You should be able to see the row of data that you inserted while DB2 was running on node ha1. The output for our run is:
[db2inst1@ha2 db2inst1]$ db2 -tf /ha/hahbcode/db2/selectdb.sql
Database Connection Information
Database server = DB2/LINUX 8.1.0
SQL authorization ID = DB2INST1
Local database alias = HADB
COL1 COL2
----------- ------------
10 Hello
1 record(s) selected.
DB20000I The SQL command completed successfully.
|
This shows that the data has survived the failover from the primary to backup.
-
Insert a row of data in the test table on ha2 node as user db2inst1:
db2 -tf /ha/hahbcode/db2/insertdb.sql.
-
You can start the primary node by simply starting heartbeat on the primary system as user root:
/etc/rc.d/init.d/heartbeat start. You should see all the services come up on the primary machine in under a minute. You can verify that DB2 is running on the primary by checking the /var/log/ha-log file and using the ps command on the backup machine.
- Inspect the contents of the test table on ha1 node as user db2inst1:
db2 -tf /ha/hahbcode/db2/selectdb.sql. You should be able to see two rows of data. The output for our run is:
[db2inst1@ha1 db2inst1]$ db2 -tf /ha/hahbcode/db2/selectdb.sql
Database Connection Information
Database server = DB2/LINUX 8.1.0
SQL authorization ID = DB2INST1
Local database alias = HADB
COL1 COL2
----------- ------------
10 Hello
10 Hello
2 record(s) selected.
DB20000I The SQL command completed successfully.
|
This shows that the data has survived failover from the backup to the primary.
|
Conclusion
Critical database applications demand a robust strategy for the preventing data loss and guaranteeing high availability of your data store. DB2 UDB makes it easy to manage large databases with excellent availability characteristics.
In this article you've seen how to easily and cost effectively implement HA for DB2 UD databases using open source software and inexpensive hardware.
Download | Description | Name | Size | Download method |
|---|
| Sample code package for this article | hahbcode.tar.gz | 25 KB | HTTP |
|---|
Resources
- Read the other articles in this series:
-
Check out the High-Availability Linux project Web site for more information on heartbeat, including downloads and heartbeat success stories.
-
Download a trial version of
IBM DB2 Universal Database Enterprise Server Edition V8.2 for Linux.
- Learn about High Availability Disaster Recovery (HADR), a new feature in DB2 UDB V8.2 that provides a high-availability solution for protection against both partial and complete site failures.
-
"Planning for Availability in the Enterprise" (developerWorks, December 2003) discusses availability and how to plan for and maintain it in an enterprise middleware environment.
-
Browse other documentation on high-availability resources from IBM.
- And while you're at it, check out these cluster and HA solutions from IBM.
- Find more resources for Linux developers in the developerWorks Linux zone.
- Get involved in the developerWorks community by participating in developerWorks blogs.
- Order the SEK for Linux, a two-DVD set containing the latest IBM trial software for Linux from DB2®, Lotus®, Rational®, Tivoli®, and WebSphere®.
- Build your next Linux development project with IBM trial software, available for download directly from developerWorks.
About the author | | | Hidayatullah H. Shaikh is a Senior Software Engineer on the IBM T.J. Watson Research Center's On-Demand Architecture and Development Team. His areas of interest and expertise include business process modeling and integration, service-oriented architecture, grid computing, e-commerce, enterprise Java, database management systems, and high-availability clusters. You can contact Hidayatullah at hshaikh@us.ibm.com. |
Rate this page
| |
