Hi Ethan, Ethan Quach wrote: > Sue, > > > 2.3.2.1 - Nit - What about sparc? Is a disk on Sparc going to have > a mock PartitionInfo just to have SliceInfo?
That was a mistake. DiskInfo has been updated to indicate that it can have either a list of PartitionInfo objects or a list of SliceInfo objects. > > 2.4.5.4 - It says these "validate" functions (this one and the one > in 2.4.4.4) act as UI helper functions; how so? If they're really > UI helpers, could a user say they really do want to only create > a partition or slice as 60G even though there is a 63G chunk > available? If so, how? Functionally, these functions will behave as they do in the GUI. 3GB of space, such as in your example, would not be shifted upward. But if there is a 63.025 GB chunk of free space, and the user punches in 63 GB, it should be adjusted up to 63.025 GB (recall that the UI only allows the user to enter up to 1/10th of a GB). That said, the example listed in the design doc is inaccurate, as the UI spec calls to reject user input of greater than the available space. The validate_partition_info function is actually not called until the user moves on to the next screen. I'll update the doc to clarify the usage of this function and provide a better example case. > > 2.7.1 - Slices 8-15 for x86 aren't really editable, are they? I found the spot in the UI spec where it indicates that 10 slices will be manipulable by the text installer: "Depending on the type of VTOC, there may be eight or ten slices." > > 2.8.1 - Just to note, the grub entry design here is going to suffer > from the same console redirection issue Alok is currently going through > for bootable AI images. Once that feature in grub is available, it > should automatically detect console redirection settings from the BIOS > and not require an explicit "console=ttyX" in the grub line, for > systems that set console redirection in the BIOS. Yes, we're aware of the issues and working on a resolution. > > 2.8.2 - Can you describe a bit more about the dependencies and > dependents of this service? (Potential) Dependents: system/console-login Dependencies: filesystem/minimal milestone/sysconfig milestone/network As mentioned above, we're still considering the best, appropriate method for gaining full control of the console. The following note is found in /var/svc/manifest/system/console-login.xml: > <!-- Note that console-login should be dependent on any services > that may need to use the console. This requirement can be met > by establishing a dependency on milestone/sysconfig which, > among other things, collects such dependencies. > --> Thanks, Keith > > > thanks, > -ethan > > > > Sue Sohn wrote: >> We have posted the design document for the Text Installer at: >> >> http://www.opensolaris.org/os/project/caiman/TextInstallerProject/design_doc_v1.4.pdf >> >> >> >> Please review and provide comments by COB Tuesday, September 15th. >> >> Thanks, >> Sue >> _______________________________________________ >> caiman-discuss mailing list >> caiman-discuss at opensolaris.org >> http://mail.opensolaris.org/mailman/listinfo/caiman-discuss > _______________________________________________ > caiman-discuss mailing list > caiman-discuss at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/caiman-discuss
