List, Just wanted to close this issue with a little more finality. I decided not to mess with the COW files right now and used the exact same script for each instance of Bering UML.
I backed up all my files in case something went wrong. I booted up two Bering UML file systems and made some changes. Saved changes. Backed up /etc. Rebooted Bering UML file systems. Checked changes. Everything seems to be working fine. If I find any bugs in this later then I will post back to the list. Regards, Eric Kiser -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Eric B Kiser Sent: Thursday, October 10, 2002 10:34 AM To: [EMAIL PROTECTED] Subject: RE: [leaf-devel] UML Questions Thanks allot Brad, After your example and rereading the documentation this makes much more sense to me the second time around. Thanks for the sanity check, Eric Kiser -----Original Message----- From: Brad Fritz [mailto:[EMAIL PROTECTED]] Sent: Wednesday, October 09, 2002 9:15 PM To: Eric B Kiser Cc: [EMAIL PROTECTED] Subject: Re: [leaf-devel] UML Questions Hello Eric, On Wed, 09 Oct 2002 20:03:47 -0400 Eric Kiser wrote: > Hi All, > > I am working with bering1.0-rc3 and using it in a UML environment. I wanting > to be able to boot up multiple bering file systems and be able to modify > them and save the changes. In the UML documentation it says that if you use > the same copy of a file system to do this then you will corrupt it. What I > am trying to figure out is how to this without one uml load writing over > another one. > > Here is my script: > > #! /bin/sh > > ./linuxuml-2.4.18-21 \ > ubd0=bering_fs \ > initrd=initrd.lrp \ > root=/dev/ram0 \ > init=/linuxrc \ > boot=/dev/ubd0:minix \ > PKGPATH=/dev/ubd0 \ > devfs=nomount \ > LRP=root,etc,local,log,modules > > If I use this script multiple times then won't I have multiple copies of > bering_fs using ubd0 and ram0? > I modified the script so that ubd0 became ubd9 and ram0 became ram9 but I > was unable to get through the boot sequence. Should this have worked? I am at the edge of (and possibly beyond) my understanding of UML here, but I believe ubd0 and ram0 are devices internal to each UML environment. If I am correct, you should be able to run multiple UML instances all using ubd0 and ram0. > Any help is appreciated. Please let me know if there is more information > needed that I have overlooked. I was able to get two copies of UML to boot simultaneously using the UML Copy-On-Write layer[1]. Here is a diff of my two start scripts: < ./linuxuml-2.4.18-45 ubd0=cow0,bering_fs \ --- > ./linuxuml-2.4.18-45 ubd0=cow1,bering_fs \ (Notice the first argument to the ubd0 parameter in each.) Aside from the different "ubd0" parameters, my start scripts are nearly identical to yours. Here's one for reference: #!/bin/sh ./linuxuml-2.4.18-45 ubd0=cow0,bering_fs \ initrd=initrd.lrp \ root=/dev/ram0 \ init=/linuxrc \ boot=/dev/ubd0:minix \ PKGPATH=/dev/ubd0 devfs=nomount \ LRP=`cat packages.minimal` I haven't done any testing beyond booting the two instances, so I am not 100% positive I didn't screw something up that would come back to haunt me later. In other words, don't blame me if the above advice causes your computer to start smoking or blow up. ;) If I am reading the howto correctly, it looks like you can also merge the COW file with its backing file into a new file system image. Good luck! --Brad [1] http://user-mode-linux.sourceforge.net/UserModeLinux-HOWTO-7.html ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ leaf-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/leaf-devel ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ leaf-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/leaf-devel
