There is a new paper, "Sharing and Maintaining SLES 11 Linux under z/VM using DCSSs and an NSS", available on a new Web page at: http://www.vm.ibm.com/linux/dcss/
In 2007, there was collaboration between IBM and Nationwide insurance that resulted in the Redpaper "Sharing and maintaining Linux under z/VM", published in Feb of '08, on the Web at http://www.redbooks.ibm.com/abstracts/redp4322.html . This introduced the concept of "read-only-root". Thanks to many, especially Steve Womer and Rick Troth on that paper. In 2009, there was collaboration between, Penn State, IBM, and later Red Hat to publish two papers, "Sharing and maintaining SLES 10 SP2 Linux under z/VM" and "Sharing and maintaining RHEL 5.3 Linux under z/VM" . They are near the top of http://linuxvm.org/present/ . These papers expounded upon the "maintenance" part of the title and upgraded to both SLES and RHEL current releases. Thanks to many again, especially Brian France, Kyle Black, and Brad Hinson on those papers. At a meeting in Endicott when the previous papers were kicked-off, a z/VM developer asked why we were using read-only LINKs to minidisks and not DCSSs. After the two papers in '09 were completed, and with help from Carlos Ordonez and Vic Cross of IBM, I have been working on the subject paper to answer the question of the Endicott developer. Thanks to the many listed in the paper. It was a challenging paper to complete, and it's not simple to implement. It does have two sections that show how to go from a conventional read-write gold master to a read-only counterpart - one shows the manual steps to convey an understanding of the process, and the other shows an automated step, with an associated bash script, to improve the speed and reliability of the process. I asked a person, for whom I have a lot of respect, to review it. He looked at it briefly and wrote: "... it is rather complicated matter, yet some of the instructions suggest that you could leave it to the novice Linux admin to follow the recipe.... I would be interested to see someone demonstrate it is worth the trouble..." So, you've been warned :)) Enjoy reading, and try to implement if you have access to the resources and you dare :)) Still, perhaps it will help us mainframers blaze a trail for Linux as a whole, using some of the tried and tested techniques that *have* proven to be worth the trouble... Of course, feedback is welcome. "Mike MacIsaac" <[email protected]> (845) 433-7061
