Creating SSI Clusters Using UML HOWTO

Brian J. Watson

           Brian.J.Watson@hp.com
        

Revision History
Revision 1.04 2002-05-29 Revised by: bjw
LDP review
Revision 1.03 2002-05-23 Revised by: bpm
LDP review
Revision 1.02 2002-05-13 Revised by: bjw
Fixed minor typos and errors
Revision 1.00 2002-05-09 Revised by: bjw
Initial release

This is a description of how to create a Single System Image (SSI) cluster of virtual User-Mode Linux (UML) machines. After explaining how to use the pre-built SSI/UML binaries, this document demonstrates what an SSI cluster can do. Then it shows more advanced users how to build their own SSI/UML kernels, ramdisks and root images. Following that, it provides an overview of how to move to a hardware-based SSI cluster. It concludes with a set of links and an invitation to contribute to the SSI Clustering project.


Table of Contents
1. Introduction
1.1. Overview of SSI Clustering
1.2. Overview of UML
1.3. Intended Audience
1.4. System Requirements
1.5. New Versions
1.6. Feedback
1.7. Copyright Information
1.8. Disclaimer
2. Getting Started
2.1. Root Image
2.2. UML Utilities
2.3. SSI/UML Utilities
2.4. Booting the Cluster
2.5. Booting an Individual Node
2.6. Crashing an Individual Node
2.7. Shutting Down the Cluster
3. Playing Around
3.1. Process Movement, Inheriting Open Files and Devices
3.2. Clusterwide PIDs, Distributed Process Relationships and Access, Clusterwide Job Control and Single Root
3.3. Clusterwide FIFOs
3.4. Clusterwide Device Naming and Access
4. Building a Kernel and Ramdisk
4.1. Getting SSI Source
4.2. Getting the Base Kernel
4.3. Applying SSI Kernel Code
4.4. Building the Kernel
4.5. Adding GFS Support to the Host
4.6. Installing the Kernel
4.7. Building GFS for UML
4.8. Building the Ramdisk
4.9. Booting the Cluster
5. Building a Root Image
5.1. Base Root Image
5.2. GFS Root Image
5.3. Getting Cluster Tools Source
5.4. Building and Installing Cluster Tools
5.5. Installing Kernel Modules
5.6. Configuring the Root
5.7. Unmounting the Root Image
5.8. Distributions Other Than Red Hat
6. Moving to a Hardware-Based Cluster
6.1. Requirements
6.2. Resources
7. Further Information
7.1. SSI Clusters
7.2. CI Clusters
7.3. GFS
7.4. UML
7.5. Other Clustering Projects
8. Contributing
8.1. Testing
8.2. Documentation
8.3. Debugging
8.4. Adding New Features
9. Concluding Remarks

    Next
    Introduction