IBM Support

IBM Business Automation Workflow (BAW) pattern V18.0.0.2 deployment succeeds but HADR setup is not correct

Release Notes


Abstract

IBM Business Automation Workflow (BAW) Db2 HADR pattern V18.0.0.2 deployment succeeds, however, HADR setup is not correct.
When db2pd -database cpedb -hadr command is run on the primary Db2 server node, the HADR_STATE is DISCONNECTED for CPEDB.
$ db2pd -database cpedb -hadr
HADR_CONNECT_STATUS = DISCONNECTED

Also, lssam command does not show HADR status of CPEDB.
$lssam
Online IBM.ResourceGroup:db2_db2inst1_db2inst1_CPEDB-rg Request=Lock Nominal=Online
'- Online IBM.Application:db2_db2inst1_db2inst1_CPEDB-rs Control=SuspendedPropagated
|- Online IBM.Application:db2_db2inst1_db2inst1_CPEDB-rs:<DB2_Primary_host_name>
'- Offline IBM.Application:db2_db2inst1_db2inst1_CPEDB-rs:<Db2_Secondary_host_name>

Content

Follow these steps to setup HADR for CPEDB and get the databases in sync:

1. On the Db2 primary node, do a complete backup of the database, online.

mkdir /tmp/cpedb (Run this command as db2inst1)

db2 backup database cpedb online to /tmp/cpedb/ (Run this command as db2inst1)

For example:

image-20191204161414-1

2. On the Standby Db2 server, copy the backup from Db2 primary to Db2 standby by using following commands:

3. On the Standby Db2 server, drop cpedb database  by using the following command:

db2 drop db cpedb (Run this command as db2inst1)

image-20191204161414-3

image-20191204161414-4

4. On the Standby Db2 server, restore the backed-up database on the standby server by using following sequence of commands (do not roll-forward it):

db2 "restore database cpedb from /tmp/cpedb dbpath on /db2data into cpedb redirect generate script restore_cpedb.sql" (Run this command as db2inst1)

image-20191204161414-5

This command creates a restore_cpedb.sql script.

5. On the Standby Db2 server, this script requires following folders to be created on the standby node. Create the folders by using the following command:

               (Run these commands as db2inst1)

  • mkdir -p /db2data/db2inst1/NODE0000/CPEDB/TOSSA/indexfs2
  • mkdir -p /db2data/db2inst1/NODE0000/CPEDB/TOSSA/indexfs1
  • mkdir -p /db2data/db2inst1/NODE0000/CPEDB/TOSSA/datafs1
  • mkdir -p /db2data/db2inst1/NODE0000/CPEDB/TOSSA/datafs2
  • mkdir -p /db2data/db2inst1/NODE0000/CPEDB/TOSSA/datafs3
  • mkdir -p /db2data/db2inst1/NODE0000/CPEDB/DOSSA/lobfs1
  • mkdir -p /db2data/db2inst1/NODE0000/CPEDB/DOSSA/indexfs1
  • mkdir -p /db2data/db2inst1/NODE0000/CPEDB/DOSSA/indexfs12
  • mkdir -p /db2data/db2inst1/NODE0000/CPEDB/DOSSA/indexfs2
  • mkdir -p /db2data/db2inst1/NODE0000/SQL00002/LOGSTREAM0000/
  • mkdir -p /db2data/db2inst1/NODE0000/CPEDB/TOSSA/lobfs1
  • mkdir -p /db2data/db2inst1/NODE0000/CPEDB/DOSSA/datafs1
  • mkdir -p /db2data/db2inst1/NODE0000/CPEDB/DOSSA/datafs2
  • mkdir -p /db2data/db2inst1/NODE0000/CPEDB/DOSSA/datafs3

6. On the Standby Db2 server, restore the database by using the following command:

db2 -tvf restore_cpedb.sql (Run this command as db2inst1)

image-20191204161414-6

7. On primary and standby nodes, stop the firewall by using following command:

            For example, on a Linux system:

systemctl stop iptables (Run this command as root)

8. On the primary node, set the HADR parameters as follows:

db2 update db cfg for cpedb using   HADR_LOCAL_HOST <Local_host_IP> HADR_LOCAL_SVC 60002 HADR_REMOTE_HOST <Remote_host_IP> HADR_REMOTE_SVC 60002 HADR_REMOTE_INST db2inst1 HADR_TIMEOUT 300 HADR_SYNCMODE nearsync (Run this command as db2inst1)

9. On the Standby Db2 server, set the HADR parameters as follows:

db2 update db cfg for cpedb using   HADR_LOCAL_HOST <Local_host_IP> HADR_LOCAL_SVC 60002 HADR_REMOTE_HOST <Remote_host_IP> HADR_REMOTE_SVC 60002 HADR_REMOTE_INST db2inst1 HADR_TIMEOUT 300 HADR_SYNCMODE nearsync (Run this command as db2inst1)

10. Start HADR on the Standby Db2 server by using the following command:

db2 start hadr on database cpedb as standby (Run this command as db2inst1)

11. Start HADR on the primary node by using the following command:

db2 start hadr on database cpedb as primary (Run this command as db2inst1)

[Optional] If lssam or db2pd -database cpedb -hadr output is still incorrect (DISCONNECTED) then stop HADR on CPEDB by using the following commands:

[Optional] – On the Standby Db2 server, stop cpedb hadr by using the following command:

db2 stop hadr on database cpedb

[Optional] – On the primary Db2 server, stop cpedb hadr by using the following command:

db2 stop hadr on database cpedb

Start HADR on standby and primary Db2 servers by using commands in the above-mentioned steps.

12. Check the HADR status of cpedb. Ensure that the status is “CONNECTED” on both, the standby and the primary Db2 servers.

$ db2pd -database cpedb -hadr

HADR_CONNECT_STATUS = CONNECTED

[{"Business Unit":{"code":"BU053","label":"Cloud & Data Platform"},"Product":{"code":"SSM8NY","label":"PureApplication System"},"Component":"","Platform":[{"code":"PF016","label":"Linux"}],"Version":"2.3.1.x;2.3.0.1;2.2.6.0","Edition":"All editions","Line of Business":{"code":"","label":""}},{"Business Unit":{"code":"BU053","label":"Cloud & Data Platform"},"Product":{"code":"SSFQSV","label":"IBM Cloud Pak System Software"},"Component":"","Platform":[{"code":"PF016","label":"Linux"}],"Version":"2.3.1.x;2.3.0.1","Edition":"All editions","Line of Business":{"code":"LOB45","label":"Automation"}}]

Document Information

Modified date:
06 May 2020

UID

ibm11119705

Page Feedback