![]() |
Home
| Calendar
| Mail Lists
| List Archives
| Desktop SIG
| Hardware Hacking SIG
Wiki | Flickr | PicasaWeb | Video | Maps & Directions | Installfests | Keysignings Linux Cafe | Meeting Notes | Linux Links | Bling | About BLU |
Yesterday I wrote: > This morning I tried switching the new SuSE hard drive into the old box. > Tried each of the 3 different settings available for block-mapping of > the drive in the BIOS, but it won't read the drive properly (crashes > before completing a kernel load, or gets a CRC error). Last night I spent some time digging through LILO documentation (after confirming that the 10-gig drive is readable as a root filesystem from a boot floppy) to see if I could figure out how to get the old P133 to boot up from the hard drive. Alas, there have been a lot of changes in how BIOS boot loaders have operated the past few years. Why don't they just come up with a way to address 1000 terabytes and be done with it; a 64-bit block number field should never be obsoleted. This BIOS dates to October 1996; it boots OK with the 2-gig drive which came with the system, but not the new one. I tried a number of LILO parameters but it always crashes about the same way (it usually finishes the kernel load but then issues a CRC or other error before splattering the screen with garbled characters). It doesn't look to me like there are any meaningful diagnostics available to tell me exactly why the kernel load fails or what parameters (other than linear or selecting LBA in the BIOS or whatever I've already tried) could be tweaked. It takes a *long* time between each attempt, so my time is really running out. (And causing domestic crises related to my failure to build shelves in my our kitchen, plus carry out other chores ;-) Before I throw in the towel and buy a brand new PC, any suggestions? All I need to get this old P133 into production, apparently, is either a successful LILO installation on the 10-gig drive, or a boot floppy which brings up /dev/hda3 from the 2.2.14 kernel which came with SuSE. (I could build a custom kernel but have no need to do so, the stock one works fine.) I dd'ed that kernel to a floppy and rdev'ed it to hda3, but that fails even worse than LILO (have no idea why; the floppy thus produced doesn't get further than "Loading" on any of my systems). -rich - Subcription/unsubscription/info requests: send e-mail with "subscribe", "unsubscribe", or "info" on the first line of the message body to discuss-request at blu.org (Subject line is ignored).
![]() |
|
BLU is a member of BostonUserGroups | |
We also thank MIT for the use of their facilities. |