2007/6/26, MC <[EMAIL PROTECTED]>:
If the zone stuff will take as much work as I think, I'd leave it off the first release feature list entirely. The Linux zone isn't really interesting because it doesn't support the 2.6 kernel, and regular zones are an advanced feature that will appeal to a tiny portion of the user base.
As long as it runs any software on top of libc, that's not an issue. Indiana will be used by curious new single users who want to check out ZFS
for their file server, or try out the new Netbeans or SunStudio, or compare it to their current OS. So you want to have the OS up to the status quo (installer, boot speed, application selection, patching) more than you want zones working.
Indiana will be used by people looking at it as a serious alternative for their linux servers. If you don't provide them the real advantages with a cherry on top then they will give it up. Anyone is free to work on this stuff of course, but I'd leave zones off the
top priority list until more commonly used features are done. Once that happens, I'd make a fuss about zones only when there is a compelling reason for normal folks to use them, and a GUI to help them do it. A possible example: putting a file server in one zone, a MythTV-like PVR in another zone, and a developer workstation in a final zone.
I agree with you in that we should start working on the inftrastructure and the base where we are going to create this zone magic thing first, and I haven't seen any real work on that yet. However, I think that we should trust our BDFL and give him some time to get the engines ready ;-) -- Un saludo, Alberto Ruiz
_______________________________________________ indiana-discuss mailing list [email protected] http://opensolaris.org/mailman/listinfo/indiana-discuss
