1. Introduction to Clusters:
Cluster (cluster) technology is a relatively new technology, through clustering technology, can be paid at a lower cost in the performance, reliability, flexibility of relatively high returns, and its task scheduling is the core technology in the cluster system.
If a thing is done by more than one person, it has a similar place to the cluster. We all share the pressure, if some people take a temporary leave, others can share his task, things will not be delayed. And when he is busy with his own things will still give him a new task, will not break the rhythm. And the cluster is in the simulation of doing such things, in fact, most of the companies are also using clusters to provide services to users. Whether it's Apache or Nginx. We do not want to browse the Web page suddenly not found it. Then you need to understand the cluster. This scheduling method and thought is very helpful for us to do everything.
2. Framework diagram of the cluster:
Here are some features of the cluster:
Heartbeat monitoring: Each cluster is connected to each other, and when the service has a problem, they disconnect the link, but all think that the other side has failed.
We are building a highly available cluster today:
1. Experimental environment:
1. System: redhat6.5
2. Host: A total of three nodes, a master control node (with a real machine) and two load nodes (with virtual machines)
Note: This highly available cluster is the software of the Redhat company, no third party software, so if you have a local Yum source on your machine, you can do it.
2. Cluster Setup steps:
(1) Our virtual machines can be made with snapshots (saving resources).
For these three nodes we have a few places to be aware of:
1. The firewall of the three is closed;
2. Three nodes parse each other (edit/etc/hosts);
3. SELinux shutdown of the control node;
4. To accurately configure the Yum source;
5. Time must be synchronized, otherwise in the process of encryption or authentication error occurred;
We can solve the above problems one by one.
(2) Control node:
First modify the Yum Source:
The following [Highavailability], [LoadBalancer], [Resilientstorage], [Scalablefielsystem] are at the mount point of the ISO, we need to manually add him up. This is the address of the above Yum:
When you are done, copy the Yum.repo file to the Yum directory of the other two load nodes:
After copying, three nodes are refreshed simultaneously:
At the control node, install the Luci software, which is the cluster control software:
The Ricci software needs to be installed at the load node:
After completion, start the service on the control node:
He'll tell you to open the Web console's network address, but before you open the Web, make sure that your control node installs the Apache service, and before we do this, we'll first modify the Ricci user's password on the load node, which will be used later.
Then open the Web page and pop up the content:
The user is root, and the password is the root password of the control node.
Clicking Create creates a cluster and pops up some pages:
Set Cluster Parameters:
This added process is lengthy, with two load nodes installing Cman, Rgmanager, Modclusterd, CLVMD services, and restarting after the service installation is complete. After the cluster has been added successfully, the following is shown:
The two-node cluster is built to do this, but this is only the most basic work, the cluster can provide high-availability services outside, the next time to explain.
Linux Advanced Techniques: Cluster building (i)
