Summary:
简单的HADR,只用一台虚拟机,两个实例间搭建。工作量不大,一般5分钟左右能够完成。
步骤:
1.设定归档模式
2.使用备份建立standby数据库
3.设定hadr相关的参数
4.启动并测试
测试环境:
OS:Red hat 6
Server: 192.168.122.17 Primary instance: hadr1pri Primary service/port: 43099 Standby instance: hadr1std Standby service/port: 44099 DB name: ORG --注意,切勿使用与DBM SVCENAME 太接近的端口,因为实例会默认使用那端口之后的连续几个端口,所以应尝试更远一些的端口
vi /etc/services
DB2_hadrpri 60008/tcp DB2_hadrpri_1 60009/tcp DB2_hadrpri_2 60010/tcp DB2_hadrpri_END 60011/tcp DB2_hadrstd 60012/tcp DB2_hadrstd_1 60013/tcp DB2_hadrstd_2 60014/tcp DB2_hadrstd_END 60015/tcp 43099 43099/tcp 44099 44099/tcp
搭建开始
step 1: 创建组和用户
create primary instance id
To create groups on Linux operating systems, enter the following commands:
groupadd -g 1999 db2iadm2 groupadd -g 1998 db2fadm2 groupadd -g 1997 dasadm2 Create users for each group: useradd -u 1014 -g db2iadm2 -m -d /home/hadrpri hadrpri useradd -u 1013 -g db2fadm2 -m -d /home/hadrfenc hadrfenc useradd -u 1012 -g dasadm2 -m -d /home/hadrdas hadrdas
create standby instance id
To create groups on Linux operating systems, enter the following commands:
useradd -u 1020 -g db2iadm2 -m -d /home/hadr1std hadr1std
useradd -u 1019 -g db2fadm2 -m -d /home/hadr1sfc hadr1sfc
useradd -u 1018 -g dasadm2 -m -d /home/std1das std1das
step 2:创建实例
cd /opt/IBM/db2/V9.7/instance ./db2icrt -s ese -u hadr1fc hadr1pri ./db2icrt -s ese -u hadr1sfc hadr1std
step3: 主节点归档模式设置,备节点使用restore方式创建数据库,主备节点HADR设置
--在Primary:
--启用归档模式 --启用LOGINDEXBUILD,以便日志有关索引的操作 db2 update db cfg for org using LOGRETAIN on db2 update db cfg for org using LOGINDEXBUILD on --Backup DB db2 backup db org to /data
----在standby
--启动数据
--使用primary db 的备份文件进行数据库恢复
db2start
db2 restore db org from /home/hadrstd taken at 20150120040252 on /home/hadrstd dbpath on /home/hadrstd
--这时候standby的数据库应该是roll-forward pedning的状态,切勿手动roll-forward
db2 connect to org
SQL1117N A connection to or activation of database "SAMPLE" cannot be made
because of ROLL-FORWARD PENDING. SQLSTATE=57019
--在Primary:
db2 update dbm cfg using svcename DB2_hadrpri
db2 update db cfg for org using hadr_local_svc 41099
db2 update db cfg for org using hadr_remote_host 192.168.122.17
db2 update db cfg for org using hadr_local_host 192.168.122.17
db2 update db cfg for org using hadr_remote_svc 42099
db2 update db cfg for org using hadr_remote_inst hadrstd
db2 update db cfg for org using hadr_syncmode sync
db2 connect to org db2 quiesce database immediate force connections db2 unquiesce database db2 connect reset
--在Standby:
db2 update dbm cfg using svcename DB2_hadrstd
db2 update db cfg for org using hadr_remote_host 192.168.122.17
db2 update db cfg for org using hadr_local_svc 42099
db2 update db cfg for org using hadr_local_host 192.168.122.17
db2 update db cfg for org using hadr_remote_svc 41099
db2 update db cfg for org using hadr_remote_inst hadrpri
db2 update db cfg for org using hadr_syncmode sync
db2 update db cfg for org using HADR_TIMEOUT 3
db2 update db cfg for org using HADR_PEER_WINDOW 120
4.启动并测试
--先启动standby
--在Standby:
db2 start hadr on db sample as standby
DB20000I The START HADR ON DATABASE command completed successfully.
--这时候应该是remote catchup pending的状态:
[hadrstd@oc0644314035 db2dump]$ db2pd -d org -hadr
Database Partition 0 -- Database ORG -- Standby -- Up 0 days 00:00:05 -- Date 2015-01-20-22.00.09.169952
HADR Information:
Role State SyncMode HeartBeatsMissed LogGapRunAvg (bytes)
Standby RemoteCatchupPending Sync 0 0
ConnectStatus ConnectTime Timeout
Disconnected Tue Jan 20 22:00:05 2015 (1421809205) 120
LocalHost LocalService
192.168.122.17 42099
RemoteHost RemoteService RemoteInstance
192.168.122.17 41099 hadrstd
PrimaryFile PrimaryPg PrimaryLSN
S0000000.LOG 0 0x0000000002728010
StandByFile StandByPg StandByLSN StandByRcvBufUsed
--再启动Primary
--在Primary:
[hadr1pri@oc0644314035 data]$ db2 start hadr on database org as primary
DB20000I The START HADR ON DATABASE command completed successfully.
[hadrstd@oc0644314035 ~]$ db2pd -d org -hadr
Database Partition 0 -- Database ORG -- Standby -- Up 0 days 00:11:53 -- Date 2015-01-21-00.59.58.732725
HADR Information:
Role State SyncMode HeartBeatsMissed LogGapRunAvg (bytes)
Standby Peer Sync 0 0
ConnectStatus ConnectTime Timeout
Connected Wed Jan 21 00:48:39 2015 (1421819319) 3
PeerWindowEnd PeerWindow
Wed Jan 21 01:01:57 2015 (1421820117) 120
LocalHost LocalService
192.168.122.17 42099
RemoteHost RemoteService RemoteInstance
192.168.122.17 41099 hadrpri
PrimaryFile PrimaryPg PrimaryLSN
S0000001.LOG 0 0x0000000002728010
StandByFile StandByPg StandByLSN StandByRcvBufUsed
S0000001.LOG 0 0x0000000002728010 0%
--可以看到一旦Primary也起来了,hadr的状态就会变成peer
--这时候尝试手动归档,看日志是否能够顺利传递到standby
[hadr1pri@oc0644314035 data]$ db2 archive log for DB sample
DB20000I The ARCHIVE LOG command completed successfully.
--在Standby进行role 切换
[hadrstd@oc0644314035 ~]$ db2 takeover hadr on database org
DB20000I The TAKEOVER HADR ON DATABASE command completed successfully.
[hadrstd@oc0644314035 ~]$ db2pd -d org -hadr
Database Partition 0 -- Database ORG -- Active -- Up 0 days 00:12:16 -- Date 2015-01-21-01.00.21.098460
HADR Information:
Role State SyncMode HeartBeatsMissed LogGapRunAvg (bytes)
Primary Peer Sync 0 0
ConnectStatus ConnectTime Timeout
Connected Wed Jan 21 00:48:39 2015 (1421819319) 3
PeerWindowEnd PeerWindow
Wed Jan 21 01:02:20 2015 (1421820140) 120
LocalHost LocalService
192.168.122.17 42099
RemoteHost RemoteService RemoteInstance
192.168.122.17 41099 hadrpri
PrimaryFile PrimaryPg PrimaryLSN
S0000001.LOG 0 0x0000000002728010
StandByFile StandByPg StandByLSN
S0000001.LOG 0 0x0000000002728010
References
搭建案例:
http://guoyanxi.iteye.com/blog/1173906
http://blog.csdn.net/dream19881003/article/details/7417285
自己总结的ISSUES
http://www.cnblogs.com/DBA-Ivan/p/4260264.html
Issues:
https://www.ibm.com/developerworks/community/forums/html/topic?id=77777777-0000-0000-0000-000014850278
http://bytes.com/topic/db2/answers/496482-unable-start-hadr-reason-code-7-a
http://www.dbforums.com/showthread.php?1665366-DB2-9-7-HADR-setup