Jay, Sorry, I'm not actually subscribed to debian-alpha, so it took me a couple of days to see this. I meant to set Mail-Followup-To: debian-boot, but that didn't seem to stick... I'll try subscribing for a while instead. :) On Wed, Jan 07, 2004 at 16:32:18 -0500, Jay Estabrook wrote: > I had no trouble along these two lines. > HOWEVER, I couldn't get the installer to find the installatiopn CD > after booting from it, regardless of IDE or SCSI CDROM drive. > Also, neither the SYM/NCR driver(s) nor the DAC960 driver were > included, making the install machine selection much smaller... > The above problems occurred with both ISOs. There were a few days' worth of bad ISO images caused by transitioning to a new kernel source package for the CDs. I think these issues are also sorted out now; yesterday's image included all of the necessary IDE and SCSI modules, and only had problems at the kernel installation stage (it grabbed a 2.6 kernel by mistake). This problem should also be fixed as of today, leaving only the lack of automatic bootloader configuration as a showstopper for beta status. That will almost certainly not be fixed in time for today's daily snapshot -- you'll get a fully working base system, but will have to install & configure aboot by hand to make it bootable. I'm working on aboot-installer now, but it looks like this is going to be tricky to get right. Unlike the i386 bootloaders, aboot requires the user to keep at least one ext2 partition on the system, and we need to let the user know about this requirement *before* installing the base system instead of after when aboot-installer runs. We're probably a few more days out from having a complete installer, but I'd certainly still appreciate testing in the meantime to catch any other problems my limited test environment missed. > And, is there any reason why there are so many IDE drivers that > couldn't possibly (IMHO) be in an Alpha, that are part of the build? I > mean, what's the sense in having a VIA mobo chipset-only IDE driver??? If you can provide some sort of authoritative list of chipsets that would never be available to an Alpha, I'd be happy to get them removed from the d-i images (as well as pushing the changes up to Herbert for the kernel-image packages in general). I certainly agree most of these IDE chipsets are unlikely to ever show up on an Alpha, but don't feel comfortable trying to say which ones it's *impossible* to find. Cheers, -- Steve Langasek postmodern programmer
Attachment:
signature.asc
Description: Digital signature