필요 소프트웨어와 버전
The software requirements for PowerHA 6.1 on 10 gR2 are as follows:
| Software | Version and other details |
|---|---|
The software requirements for PowerHA 6.1 on 11 gR1 are as follows:
| Software | Version and other details |
|---|---|
The software requirements for PowerHA 6.1 on 11 gR2 are as follows:
| Software | Version and other details |
|---|---|
오라클 shared storage option
Oracle RAC 10g ~ 11gR1
| Storage option | OCR and vote disk | Oracle Software | Oracle database files |
|---|---|---|---|
Oracle RAC 11gR2
| Storage option | OCR and v ote disk | Oracle Software | Oracle database f iles |
|---|---|---|---|
Memory and Paging space Requirements
- Memory
1.5GB for Oracle Grid Infrastructure
2.5GB for Oracle Grid Infrastructure and oracle RAC - Swap space
Minimum : 1.5GB
System with 2GB of RAM or less : 1.5 times the amount of RAM
System with 2GB to 16GB of RAM : equal to RAM
system with more than 16GB of RAM : 16GB
Consideration on OCR and vote devices
- With normal redundancy
3 OCR and 3 vote devices are needed
each device needs at least 280MB in size - OCR and vote on raw devices are not supported from 11gR2
ASM or cluster file system are supported
Operating System Requirements
- Operating system
- AIX 7.1 Base
- AIX 6.1 TL02 SP1
- AIX 5.3 TL09 SP1 - Operting system filesets(for AIX6.1)
openssh
bos.adt.base
bos.adt.lib
bos.adt.lib
bos.perf.libperfstat
bos.perf.perfstat
bos.perf.proctools
rsct.basic.rte
rsct.compat.clients.rte
xlC.aix61.rte 10.1.0.0
Name Resolution
#cat /etc/hosts
127.0.0.1 loopback localhost
130.130.10.115 node15
130.130.10.116 node16
192.168.10.115 node15rac
192.168.10.116 node16rac
130.130.10.215 node15vip
130.130.10.216 node16vip
130.130.10.15 node15scan
- Scan IP(Single client access name)
- Host name to provied for all clients connecting to the cluster
- The SCAN is a domain name registered to at least one and up to three IP addresses, either in the domain name service(DNS) or the Grid Naming Service(GNS)
- The SCAN eliminates the need to change clients when nodes are added to or removed from the cluster
Configure OS kernel parameters
- Tuning Virtual Memory Manager
#vmo -p -o minperm%=3
#vmo -p -o maxperm%=90
#vmo -p -o maxclient%=90
#vmo -p -o lru_file_repage=0
#vmo -p -o strict_maxclient=1
#vmo -p -o strict_maxperm=0 - From AIX6.1, these are all default values
Configure AIX System Environment
| Shell Limit | Item in limits.conf | Hard Limit |
|---|---|---|
/etc/security/limits
default :
fsize = -1
data = -1
core = -1
fsize_hard = -1
cpu_hard = -1
data_hard = -1
stack_hard = 65536
core_hard = -1
Tuning AIX System Environment
Configuring Network Tuning Parameters
Parameter Value
#no -p -o udp_sendspace=65536
#no -p -o udp_recvspace=655360
#no -p -o tcp_sendspace=65536
#no -p -o tcp_recvspace=65536
#no -p -o rfc1323=1
#no -p -o sb_max=4194304
#no -r -o ipqmaxlen=512
Users and Groups
- Configuring users
oragrid : create a user for installing the Oracle Grid Infrastructure software
oracle : user for installing the Oracle RAC software and creating database
From 11g, run the following command as root
#chuser capabilities=CAP_NUMA_ATTACH, CAP_BYPASS_RAC_VMM, CAP_PROPAGATE oragrid
#chuser capabilities=CAP_NUMA_ATTACH, CAP_BYPASS_RAC_VMM, CAP_PROPAGATE oracle
#lsuser -a capabilities oragrid (checking capabilities for oragrid)
- Configuring groups
dba : primary group for the oragrid and oracle user
hagsuser : Required in PowerHA environment
Network Time Protocol
- NTP in AIX
시간 동기화 서버 환경구성
- edit /etc/ntp.conf
server 127.127.1.0 prefer 추가(자신을 동기화 함)
- stop ntpd and restart it with the -x option
#startsrc -s xntpd -a "-x"
- edit /etc/rc.tcpip to add the -x option to xntpd on all nodes
start /usr/sbin/xntpd "$src_running" "-x"
Note : -x option prevents time from being adjusted backward(Time backward를 막는다. 시간이 늦는 서버의 시간을 동기화 하는 것을 방지)
- lssrc -ls xntpd 결과 값 확인root@v5:/>lssrc -ls xntpd
Program name: /usr/sbin/xntpd
Version: 3
Leap indicator: 00 (No leap second today.)
Sys peer: 127.127.1.0
Sys stratum: 4
Sys precision: -16
Debug/Tracing: DISABLED
Root distance: 0.000000
Root dispersion: 0.010010
Reference ID: 127.127.1.0
Reference time: c52992e1.7c721000 Tue, Oct 26 2004 22:22:09.486
Broadcast delay: 0.003906 (sec)
Auth delay: 0.000122 (sec)
System flags: pll monitor filegen
System uptime: 73593 (sec)
Clock stability: 0.000000 (sec)
Clock frequency: 0.000000 (sec)
Peer: 127.127.1.0
flags: (configured)(refclock)(sys peer)(preferred)
stratum: 3, version: 3
our mode: client, his mode: server
Subsystem Group PID Status
xntpd tcpip 352262 active
시간동기화 Client 환경 구성
- xntpd는 Server와 Client 간 1000초(16분) 이상의 차이가 발생하면 더 이상 시간을 동기화 하지 않는다. date로 확인 후 시간차가 크면 근접한 시간으로 변경
- edit /etc/ntp.conf
server (server_name or ip)
- edit /etc/rc.tcpip
start /usr/sbin/xntpd "$src_running""-x" 주석 제거
- xntpd 기동
#startsrc -s xntpd -a "-x"
-#lssrc -ls xntpd 결과값 확인root@v4:/> lssrc -ls xntpd
Program name: /usr/sbin/xntpd
Version: 3
Leap indicator: 11 (Leap indicator is insane.)
Sys peer: no peer, system is insane
Sys stratum: 16
Sys precision: -16
Debug/Tracing: DISABLED
Root distance: 0.000000
Root dispersion: 0.000000
Reference ID: no refid, system is insane
Reference time: no reftime, system is insane
Broadcast delay: 0.003906 (sec)
Auth delay: 0.000122 (sec)
System flags: pll monitor filegen
System uptime: 23 (sec)
Clock stability: 0.000000 (sec)
Clock frequency: 0.000000 (sec)
Peer: v5.kr.ibm.com
flags: (configured)(preferred)
stratum: 4, version: 3
our mode: client, his mode: server
Subsystem Group PID Status
xntpd tcpip 15002 active
- ntpq -d로 확인root@v4:/> ntpq -p
remote refid st t when poll reach delay offset disp
========================================================================
*v5.kr.ibm.com LOCAL(0) 4 u 35 64 377 1.27 -3.345 0.53
refid 가 LOCAL로 바뀌고 reach 가 377에 다다르면 정상(6분~7분)
- Oracle CTSS
- Installed in active mode if NTP daemon in AIX is not active
- Started in observer mode if NTP is active
- In order to confirm that ctssd is active, run the following command
$crsctl stat resource ora.ctssd -t -init
Asynchronous I/O
- AIX 6.1 or later
- AIO device drivers are enabled by default
- Increase the number of aioservier processes to 65536
#ioo -p -o aio_maxreqs=65536 - AIX 5.x
- rootpre.sh enables AIO device driver
- use smit aio change the maxreqs parameter
SSH configuration
- node1 서버에서 /.ssh로 이동하여 #ssh-keygen -t dsa 명령으로 id_dsa, id_dsa.pub 생성
- id_dsa.pub를 node2 서버의 /.ssh로 authorized_keys라는 이름으로 copy (rcp나 scp를 이용함. 방법은 여러가지)
- node2 서버에서도 위와 같은 작업 수행
- ssh로 접속 시 패스워드 물어 보지 않고 바로 접속 가능
GPFS 설정
'IBM > HACMP' 카테고리의 다른 글
| HACMP_PD교육 (0) | 2012.11.09 |
|---|---|
| 바뀐LUN사이즈를VG에적용할때 (0) | 2012.11.09 |
| PowerHA7.1교육 (0) | 2012.11.09 |
| How to synchronize timestamp (0) | 2012.11.09 |
| zero-off-set (0) | 2012.03.24 |
| Understanding_active_and_passive_varyon_in_enhanced_concurrent_mode (0) | 2011.08.24 |
| Fast_disk_takeover (0) | 2011.08.24 |
| Enhanced_concurrent_mode (0) | 2011.08.24 |
| clstat 실행 시 error 날 때 해결책 (0) | 2011.07.21 |
| PowerHA 7.1 vs Veritas Cluster erver (VCS) (0) | 2011.07.20 |