Linux HA Cluster High-availability cluster HEARTBEAT2

First, elaborated the Linux HA cluster use background;

1.1 High-availability cluster definitions:

Highly available cluster full name: High Availability Cluster, simply put, the cluster is a set of highly scalable, high-availability, cost-effective computers. They provide the user with a set of network resources as a whole. One of the individual computer systems is a cluster node. The main function of high-availability cluster software is to automate fault checking and business switching to provide uninterrupted service.

1.2 Key benefits of the cluster system:

(1) High scalability:
(2) Ha: A node in a cluster fails, and its tasks can be passed to other nodes. Can effectively prevent single point of failure.
(3) High performance: load Balancing cluster allows the system to access more users at the same time.
(4) Cost-effective: high-performance systems can be constructed with inexpensive, industry-standard hardware.

Specific HA measurement criteria:

99% downtime of less than 4 days a year

99.9% downtime of less than 10 hours a year

99.99% downtime of less than 1 hours a year

99.999% downtime less than 6 minutes a year

Second, summarize the system composition of Linux ha cluster, the working model of HA cluster;

Third, through the lamp combination, based on heartbeat V2 CRM implementation ha; requirements, the deployment of WordPress, for editing any data in the article can be accessed after the node switch;

1. Prerequisites for configuring HA Clusters:

⑴ time between nodes must be synchronized: using the NTP protocol implementation;

Two node server node1.duoyoumi.com:192.168.1.193 and node2.duoyoumi.com:192.168.1.194 sub-installation NTP service, after installation, two nodes synchronize a stable time server time, Currently on the node1.duoyoumi.com server

Ntpdate 192.168.1.187;ssh 192.168.1.194 ' ntpdate 192.168.1.187 '

⑵ Inter-node requirements through the host name to communicate with each other, must resolve the host to the IP address;

node1.duoyoumi.com 192.168.1.193 Node 1 lamp+heartbeat2

node2.duoyoumi.com 192.168.1.194 Node 2 lamp+heartbeat2

node3.duoyoumi.com 192.168.1.195 Node 3 NFS shared storage

⑶ consider whether the arbitration equipment is used;

⑷ establish the root user between the nodes can be based on key authentication;

Key authentication is the other system generated key pair of the public key appended to the local system in the/root/.ssh/authorized_keys file, the local system can implement the key authentication method login, conversely, remember to append, or else may be the other server key to cover the

2. Package Installation:

⑴ Dependent Package Installation: Yum-install net-snmp-libs libnet PyXML perl-time-date

⑵heartbeat Package Installation: RPM-IVH heartbeat-2.1.4-12.e16.x86_64.rpm heartbeat-pils-2.1.4-12.e16.x86_64.rpm heartbeat- stonith-2.1.4-12.e16.x86_64.rpm

The general error in the process is libnet package through Yum can not be installed, you have to manually next Libnet binary package installation

3. Modify the configuration file:

After the heartbeat package is successfully installed, first copy the HA.CF, Authkeys, and haresources three files from the/usr/share/doc/heartbeat-2.1.4 to the node server/etc/ HA.D directory, note: Authkeys file permissions must be 600 or 400

⑴HA.CF: Master configuration file, defining Heartbeat ha basic properties on each node

    # Debugfile/var/log/ha-debug whether the debug feature is turned on, the file location for which the feature has been debugged (you can choose this option)              logfacility     local0     Logging with Rsyslog and pointing to local0

                                  &N Bsp                       

keepalive 1    Heartbeat, replacement time

Warntime 8 warning Time

Initdead How long after the heartbeat starts to check the heartbeat

Udpport 694 Heartbeat Port 694, after starting heartbeat, can be viewed by ss-utnl

#bcast eth0 # Linux Linux broadcast ports

mcast eth0 225.23.190.1 694 1 0 multicast address, can be ifconfig to see if a multicast address is opened

650) this.width=650; "src=" http://s3.51cto.com/wyfs02/M00/74/E9/wKioL1YuJkngoMClAAJTzk89WP8970.jpg "title=" 123. PNG "alt=" wkiol1yujkngomclaajtzk89wp8970.jpg "/> If not, open the multicast mode via IP link set eth0 multicast on

#ucast eth0 192.168.1.2 Multicast address

Auto_failback on fault auto recovery, default on

Node node1.com nodes, remember that the IP address cannot be used in this office and the name must be the same as the hostname result

Node node2.com nodes, remember that you cannot use an IP address in this office

#compression bz2 compression transfer algorithm on the primary node start whether to automatically switch back to the master node

ping 192.168.1.187 quorum device, can point to Gateway

#ping_group group1 10.10.10.25410.10.10.253 Quorum device Group

#debug levels of 1debug

Minimum size of #compression_threshold 2 compression, in kilobytes

CRM on open CRM also disables Haresource.

⑵authkeys: The encryption algorithm and key are used when passing messages to each other in the cluster; file permissions must be set to 600

Edit the Authkeys file to generate a character password using OpenSSL random-base64 23

[email protected] ha.d]# OpenSSL rand-base64 23
Hgtjsdpdwvtvb+6okn7mfnbokglqhhe=
# Vim/etc/ha.d/authkeys
Auth 2
#1 CRC
2 SHA1 hgtjsdpdwvtvb+6okn7mfnbokglqhhe=
#3 MD5 hello!
Wq
[[email protected] HA.D] #chmod Authkeys

⑶ Installing lamp

Yum install http mysql-server php php-mysql

After successful installation

This article is from the "Multi-Surveyors" blog, please be sure to keep this source http://icestick8586.blog.51cto.com/9867871/1834098

Linux HA Cluster High-availability cluster HEARTBEAT2