Re: WARNING: the latest boot floppies are broken
Hi,
On Tue, 11 Jan 2000, Ben Collins wrote:
> a) busybox's init seems to be broken. If you switch to vt 2 (ALT+F2) you
> can hit enter and manually start `dbootstrap'
Dbootstrap seems to start just fine on my SparcStation SLC.
> b) the filename selection for drivers.tgz and rescue.bin when installing
> those from mounted or nfs is broken. This is hardcoded, so wait till the
> next upload tonight.
I quick-fixed it on vt2 (mkdir /target/tmp/sun4c{,/disks-1.44)} and could
continue the install.
> I have succesfully booted via nfsroot and installed on to a mounted nfs
> partition (dbootstrap wont let you select an NFS partition for the root fs
> to install to, you have to manually mount it via the shell).
IIRC it worked for me though.
> This is on
> ultrasparc, so let me know about others. Note, I plan to remove RARP from
> the kernels, and force putting all the info on the boot command, simply
> because kernel RARP detects the IP wrongly (and PROM detects it
> correctly).
RARP works absolutely fine here (SLC).
Cheers,
Joost
Re: WARNING: the latest boot floppies are broken
On Tue, Jan 11, 2000 at 10:34:52PM +0100, Eric Delaunay wrote: > Ben Collins wrote: > > This is in reference to the boot floppies I just posted about earlier > > today. > > > > a) busybox's init seems to be broken. If you switch to vt 2 (ALT+F2) you > > can hit enter and manually start `dbootstrap' > > > > b) the filename selection for drivers.tgz and rescue.bin when installing > > those from mounted or nfs is broken. This is hardcoded, so wait till the > > next upload tonight. > > > > I have succesfully booted via nfsroot and installed on to a mounted nfs > > partition (dbootstrap wont let you select an NFS partition for the root fs > > to install to, you have to manually mount it via the shell). This is on > > ultrasparc, so let me know about others. Note, I plan to remove RARP from > > the kernels, and force putting all the info on the boot command, simply > > because kernel RARP detects the IP wrongly (and PROM detects it > > correctly). > > What is the problem with RARP ? I always make use of it to boot my > sparc2 & SparcClassic, but might only be a problem on ultra ? > Anyway, we don't expressly need it for sun4m & sun4u kernels as the more > common install would be based on tftpboot image that don't required it. > Could you leave it in sun4c kernel for very old sparcstations that don't have > enough memory to fit tftpboot image in and then use RARP+NFSROOT method. Well, the PROM RARP works perfectly fine, and you can specify the IP on the boot line at the PROM prompt. However, when the kernel RARP kicks it, it gets the wrong address, not sure why. -- ---===-=-==-=---==-=-- / Ben Collins -- ...on that fantastic voyage... -- Debian GNU/Linux \ ` [EMAIL PROTECTED] -- [EMAIL PROTECTED] -- [EMAIL PROTECTED] ' `---=--===-=-=-=-===-==---=--=---'
Re: WARNING: the latest boot floppies are broken
Ben Collins wrote: > This is in reference to the boot floppies I just posted about earlier > today. > > a) busybox's init seems to be broken. If you switch to vt 2 (ALT+F2) you > can hit enter and manually start `dbootstrap' > > b) the filename selection for drivers.tgz and rescue.bin when installing > those from mounted or nfs is broken. This is hardcoded, so wait till the > next upload tonight. > > I have succesfully booted via nfsroot and installed on to a mounted nfs > partition (dbootstrap wont let you select an NFS partition for the root fs > to install to, you have to manually mount it via the shell). This is on > ultrasparc, so let me know about others. Note, I plan to remove RARP from > the kernels, and force putting all the info on the boot command, simply > because kernel RARP detects the IP wrongly (and PROM detects it > correctly). What is the problem with RARP ? I always make use of it to boot my sparc2 & SparcClassic, but might only be a problem on ultra ? Anyway, we don't expressly need it for sun4m & sun4u kernels as the more common install would be based on tftpboot image that don't required it. Could you leave it in sun4c kernel for very old sparcstations that don't have enough memory to fit tftpboot image in and then use RARP+NFSROOT method. Regards. -- Eric Delaunay | "La guerre justifie l'existence des militaires. [EMAIL PROTECTED] | En les supprimant." Henri Jeanson (1900-1970)
Re: ANN: initial sparc potato boot floppies ready for testing
J. S. Connell wrote: > On Sun, 9 Jan 2000, Eric Delaunay wrote: > > > On this topic, do you know about a way of identifying the type of console > > connected to the serial line ? Is there an escape sequence that returns > > vt100 > > or the like ? > > [...] > > I'd tend to say that assuming VT100 until told otherwise by the user isn't > a bad idea. In fact, you can probably get away with assuming VT102. Thanks for all the messages I received. I will try to add code to init.c that will assume TERM=vt102 by default when booting on serial console. It could be overriden by passing TERM=xxx at boot time. However, I don't have time to write it immediatly. Just wait few more days... will go in before next monday. Btw, there are no vt100/vt102 terminfos in boot-floppies but it does apparently not hurt dbootstrap. Do I need to add them to the terminfo database excerpt the boot-floppies have in? Regards. -- Eric Delaunay | S'il n'y a pas de solution, c'est qu'il n'y [EMAIL PROTECTED] | a pas de problème. Devise Shadok.
WARNING: the latest boot floppies are broken
This is in reference to the boot floppies I just posted about earlier today. a) busybox's init seems to be broken. If you switch to vt 2 (ALT+F2) you can hit enter and manually start `dbootstrap' b) the filename selection for drivers.tgz and rescue.bin when installing those from mounted or nfs is broken. This is hardcoded, so wait till the next upload tonight. I have succesfully booted via nfsroot and installed on to a mounted nfs partition (dbootstrap wont let you select an NFS partition for the root fs to install to, you have to manually mount it via the shell). This is on ultrasparc, so let me know about others. Note, I plan to remove RARP from the kernels, and force putting all the info on the boot command, simply because kernel RARP detects the IP wrongly (and PROM detects it correctly). -- ---===-=-==-=---==-=-- / Ben Collins -- ...on that fantastic voyage... -- Debian GNU/Linux \ ` [EMAIL PROTECTED] -- [EMAIL PROTECTED] -- [EMAIL PROTECTED] ' `---=--===-=-=-=-===-==---=--=---'
my scsi hard disk doesn't work with my cdrom
Hello I need help installing a new scsi hard disk in an old sparcstation 5. For some reason, I can not get the drive to work properly with the existing cdrom in the sparcstation. Here are the facts: The sparcstation had an older, smaller, slower scsi hard disk which worked fine with the cdrom. The new harddisk works, but it causes the cdrom to disconnect from the scsi bus during large file copies from the cdrom to the hard disk. That is, 90% of the time when I copy large files from the cdrom to the hard disk, the copy will fail midway and I will be dumped back to the shell prompt. At the shell prompt, it appears as if the cdrom is gone. However, "mount" shows the cdrom as still mounted. If I "umount" the cdrom and then re"mount" it, it appears ok and I use it again for small file transfers. The problem exists using Debian Linux or OpenBSD. I've tried various jumper settings on the cdrom and on the hardisk. The settings on the old and new harddisks are not identical though. See the attached URLs. I checked http://docs.sun.com for the settings on the cdrom. It says that only the scsi id and termination jumpers should be installed. (I added the parity jumper as well.) Finally, I'm beginning to wonder if the cdrom is bad. Is there anyway I can test it? Relevent links: The new hard disk: http://www.storage.ibm.com/techsup/hddtech/dchs/dchsjum.htm The old hard disk: http://www.seagate.com/support/disc/specs/scsi/st31200w.html John Davis
Re: ANN: new boot disks to test
> Changes: > > * The sun4c kernel now has BOOTP/RARP/NFSROOT in the kernel. The sparc > and sun4u disks have RARP/NFSROOT support. > * Lots of general boot-floppies changes > * The sun4c a.out image is now padded correctly. * Lots of sparc related documentation changes in the doc stuff. Please read through this and suggest any changes/additions/removals. -- ---===-=-==-=---==-=-- / Ben Collins -- ...on that fantastic voyage... -- Debian GNU/Linux \ ` [EMAIL PROTECTED] -- [EMAIL PROTECTED] -- [EMAIL PROTECTED] ' `---=--===-=-=-=-===-==---=--=---'
ANN: new boot disks to test
Ok, at http://xia01.kachinatech.com/~buildd/sparc-2.2.14-potato-boot/current you will find the latest set of boot disks. Changes: * The sun4c kernel now has BOOTP/RARP/NFSROOT in the kernel. The sparc and sun4u disks have RARP/NFSROOT support. * Lots of general boot-floppies changes * The sun4c a.out image is now padded correctly. The images at http://xia01.kachinatech.com/~buildd/ are the ones I used to build these. I am really interested in some more sun4u and sun4c testing. Any one who had problems before, please let me know how these do. Ben -- ---===-=-==-=---==-=-- / Ben Collins -- ...on that fantastic voyage... -- Debian GNU/Linux \ ` [EMAIL PROTECTED] -- [EMAIL PROTECTED] -- [EMAIL PROTECTED] ' `---=--===-=-=-=-===-==---=--=---' pgpOR2Kx7w45B.pgp Description: PGP signature
where is kernel src and after install problems?
Hi all, I've seen kernel 2.2.14 header .deb pkg in Ben Collins' location. where is, if already available kernel 2.2.14 src .deb pkg? I need it to build a smp kernel on SS10 dual 390z50 CPU system. Despite my former email i found some problems after having installed from diskette: trying to use man command got message command not found but the man-db pkg is installed. I had to reinstal from .deb pkg. I also had to reinstall groff before. Trying to install dpkg-dev or db-make I got unresolved dependancies errors and I had to reinstall perl. After having reinstalled the pkgs with dpkg everything I used seems fine. Why this? thanks Per Lundberg for your advice on g++ and lib*, I used for some else situations like that. Infinitamente gratis, Infinitamente meglio http://www.infinito.it
