Notes for installing Oracle 11g RAC

Notes for installing Oracle 11g RAC

This article is only used to record some issues that occur during the installation of Oracle RAC at ordinary times. If you encounter minor issues, we will also add them.

1. HAIP hosts file dual-inline Nic configuration example:

# Vi/etc/hosts
127.0.0.1 localhost. localdomain localhost
192.168.0.6 db1
192.168.0.7 db2
192.168.0.8 db1-vip
192.168.0.9 db2-vip
10.10.10.10 db1-priv1
20.20.000010 db1-priv2
10.10.10.20 db2-priv1
Megaamount 000020 db2-priv2

2. Configure multi-nic binding (optional)

Note:
(1) bind two ports to the same vswitch, and configure the server as follows, prompting you to load bonding to the kernel first.
(2) If the two NICs on the server are connected to two different vswitches respectively, they can only use the Master/Slave Mode. I personally feel that there are unstable factors, such as the need to manually configure the gateway.
(3) The two NICs on the server are connected to one vswitch respectively. If the vswitch binding mode is not configured, only the active/standby mode can be used.
(4) Oracle 11.2.0.2 RAC and later versions support a maximum of four inline NICs to improve performance. You do not need to bind them. You can deploy multiple NICs as appropriate.

3. oracle 11g RAC iptables configuration, originally the database is generally not configured with a firewall, but various factors require configuration, it is difficult to find a foreign Article configuration, after installation practice and application no problem.

Iptables-P OUTPUT ACCEPT
Iptables-a input-I lo-j ACCEPT
Iptables-a input-p icmp-m icmp -- icmp-type 8-j ACCEPT
Iptables-a input-p tcp-m tcp -- dport 22-j ACCEPT
Iptables-a input-m state -- state NEW-m tcp-p tcp-s 192.168.0.0/24 -- dport 1521-j ACCEPT
Iptables-a input-p all-m state -- state ESTABLISHED, RELATED-j ACCEPT
Iptables-a input-s 192.168.0.6/32-I eth0-j ACCEPT
Iptables-a input-s 192.168.0.7/32-I eth0-j ACCEPT
Iptables-a input-s 192.168.0.8/32-I eth0-j ACCEPT
Iptables-a input-s 192.168.0.9/32-I eth0-j ACCEPT
Iptables-a input-s 192.168.0.10/32-I eth0-j ACCEPT
Iptables-a input-I eth1-j ACCEPT
Iptables-a input-I eth2-j ACCEPT
Iptables-a input-s 10.10.10.0/24-d 230.0.1.0-j ACCEPT
Iptables-a input-s 10.10.10.0/24-d 224.0.0.20.- j ACCEPT
Iptables-a input-s ACCEPT limit 20.0/24-d 230.0.1.0-j ACCEPT
Iptables-a input-s ACCEPT limit 20.0/24-d 224.0.0.20.- j ACCEPT
Iptables-a input-I eth0-d 230.0.1.0-j ACCEPT
Iptables-a input-I eth0-d 224.0.0.20.- j ACCEPT
Iptables-a input-p all-m state -- state INVALID, NEW-j DROP
Iptables-P INPUT DROP

Service iptables save

4. For Linux 6.x platform, you may find that modifications to the nproc parameter in the limits. conf file are ignored.

# Vi/etc/security/limits. d/90-nproc.conf

# Modify it to the following content to implement nproc restrictions:

* Soft nproc 16384
Root soft nproc unlimited

5. Disable THP

On the RedHat6, OEL6, SLES 11, or UEK2 kernel, make sure to disable THP (Transparent HugePages) to prevent the node and instance from being evicted due to performance issues.

Reference:
ALERT Disable Transparent HugePages on SLES11 RHEL6 OEL6 and UEK2 Kernels (Doc ID 1557478.1)

# Vi/etc/rc. local

Echo never>/sys/kernel/mm/transparent_hugepage/enabled
Echo never>/sys/kernel/mm/transparent_hugepage/defrag

6. Configure ESXi5.1 shared storage

VMware ESXi 5.1 installs RHEL6.5 11g rac and uses virtual machine disks as shared storage disks. Notes.
The SCSI controller selects LIS Logic for parallelism, And the type selects "virtual". The "virtual" disk can be shared on the same server, and the virtual device node of the disk selects, and so on, 6 shared disks are added here.
After node1 creates a disk, node2 selects an existing disk and select the path to add it.
Add the following statement to the vmx configuration files of node1 and node2. Otherwise, the disk UUID cannot be obtained.
Disk. enableUUID = "TRUE"
Note: In this environment, I personally feel that the stability is not high, and it is okay to use the test environment to build and use it. I have encountered the problem that the shared storage method is not used for reliability.

7. Example of the udev configuration file for the ASM disk in the udev mode of the EMC Storage

The format of the udev file for EMC storage configuration is different. At least it is different from that for IBM storage. Therefore, it must be recorded in/etc/udev/rules. the 50-udev.rules file is added to the d directory. This file exists under redhat 5 and is no longer available under redhat 6, and the following information is added to 50-udev.rules. Now asm can be installed normally on oracle.
Vi/etc/udev/rules. d/50-udev.rules

SUBSYSTEM = "block", KERNEL = "emcpowera1", GROUP = "asmadmin", OWNER = "grid", MODE = "0660"
SUBSYSTEM = "block", KERNEL = "emcpowerb1", GROUP = "asmadmin", OWNER = "grid", MODE = "0660"
SUBSYSTEM = "block", KERNEL = "emcpowerc1", GROUP = "asmadmin", OWNER = "grid", MODE = "0660"
SUBSYSTEM = "block", KERNEL = "emcpowerd1", GROUP = "asmadmin", OWNER = "grid", MODE = "0660"
SUBSYSTEM = "block", KERNEL = "emcpowere1", GROUP = "asmadmin", OWNER = "grid", MODE = "0660"
SUBSYSTEM = "block", KERNEL = "emcpowerf1", GROUP = "asmadmin", OWNER = "grid", MODE = "0660"
SUBSYSTEM = "block", KERNEL = "emcpowerg1", GROUP = "asmadmin", OWNER = "grid", MODE = "0660"
SUBSYSTEM = "block", KERNEL = "emcpowerh1", GROUP = "asmadmin", OWNER = "grid", MODE = "0660"
SUBSYSTEM = "block", KERNEL = "emcpoweri1", GROUP = "asmadmin", OWNER = "grid", model = "0660"
SUBSYSTEM = "block", KERNEL = "emcpowerj1", GROUP = "asmadmin", OWNER = "grid", model = "0660"
SUBSYSTEM = "block", KERNEL = "emcpowerk1", GROUP = "asmadmin", OWNER = "grid", MODE = "0660"