David Cantrell <david@slackware.com> wrote:
> In case you didn't notice, I've dumped a lot of stuff into the SPARC tree.
> Mutt, new Pine, OpenSSL, OpenSSH, XFree86 4.0.1, KDE 2.0.1, and the list
> goes on. This stuff needs testing!
>
> Please, for the love of god, someone tell me if they've successfully
> installed the latest snapshot or not! :)
The fdisk program is definitely bugged:
Command (m for help): n
Partition number (1-8): 4
First sector (2841334-4187232): 2841336
Sector 684 is already allocated
First sector (2841334-4187232):
Value out of range.
First sector (2841334-4187232):
Value out of range.
First sector (2841334-4187232):
Value out of range.
First sector (2841334-4187232):
Value out of range.
First sector (2841334-4187232): 2841334
Sector 684 is already allocated
First sector (2841334-4187232): 3000000
Sector 723 is already allocated
First sector (2841334-4187232): 4187200
Sector 1008 is already allocated
First sector (2841334-4187232):
Due to bugs I've also found in the Intel version, I've pretty much
decided I need to write another one. Others like cfdisk and sfdisk
don't do what I want for some other projects. Now I just need to
find the time to develop it.
I'm still having troubles getting the CDROM read on the Sparc side.
And I don't have a place to run NFS (yet) on the Intel side. So
what I tried to do was set up a partition on /dev/sdb to copy all
the files from the CD and just install from there. When I got to
the point of running tar to tar, there were I/O errors reading from
the CD. This may be what is causing setup to not find the CDROM.
My workaround was to dd the CD image to /dev/sdb1, mount that,
mkfs /dev/sdb4, mount that, then tar copy from sdb1 to sdb4.
I didn't know if the mount in setup would handle sdb4 being in
ISO9660 format, so I did it this way.
The interesting thing is, there are CDROM I/O errors when reading
a mounted CD, but not when reading the device image directly.
Now I get these error messages (garbled characters may be from the
serial port or from minicom):
| You can't run pkgtool from the rootdisk until you've mounted your Linux
| partitions beneath /mnt. Here are some examples of this:
|
| If yoour root partitn is /dev/hda1, and is using ext2fs, you would type:
| mount /dev/hda1 /mnt -t ext2
|
| Then, supposing your /usr partition is /dev/hda2, you must do this:
| mount /dev/hda2 /mnt/usr -t ext2
|
| Pulease mount youux partitions and then run pkgtool again.
|
| /usr/lib/setup/setup: cannot create /mnt/etc/fstab: directory nonexistent
| /usr/lib/setup/setup: cannot create /mnt/etc/fstab: directory nonexistent
| /usr/lib/setup/setup: cannot create /mnt/etc/fstab: directory nonexistent
| /usr/lib/setup/setup: cannot create /mnt/etc/fstab: directory nonexistent
| /usr/lib/setup/setup: cannot create /mnt/etc/fstab: directory nonexistent
I'm suspecting that some shell script commands in setup are failing
because they have ">/dev/tty4" on them, and the mount command is
one that is affected.
I made symlinks from /dev/tty1 ... /dev/tty4 -> /dev/null and now the
various actions take longer, so maybe they were never executing in
the first place. For example, formatting swap space before was taking
about 1 second, but now it is taking much longer. Formatting other
stuff is similarly longer.
Now it is installing series A.
I got a couple error messages during install of "etc" saying md5sum not
found. Maybe some package scripts expect md5sum. I know I would find it
useful sometimes. The "sun4u" package just had the same errors.
Series A now seems to be installed. SILO didn't seem to set the PROM boot
device, and I forget the syntax at the moment to describe the disk. I went
ahead and boot the CDROM kernel and specified root=/dev/sda1. Below is a
sampling of some error messages starting near the end of the ones regarding
the root not being mounted read-only:
| If you boot from a bootkernel disk, or with Loadlin, you can add the 'ro' flag.
|
| This will fix the problem *AND* eliminate this annoying message. :^)
|
| Press ENTER to continue.
| none on /dev/pts type devpts (rw,gid=5,mode=620)
| none on /proc type proc (rw)
| Setting system time from hardware clock...
| Updating module dependencies for Linux 2.2.17:
| depmod: ELF file not for this architecture
| depmod: ELF file not for this architecture
| depmod: ELF file not for this architecture
| depmod: ELF file not for this architecture
| depmod: ELF file not for this architecture
| depmod: ELF file not for this architecture
| depmod: ELF file not for this architecture
| depmod: ELF file not for this architecture
| depmod: ELF file not for this architecture
| depmod: ELF file not for this architecture
| depmod: ELF file not for this architecture
| depmod: ELF file not for this architecture
| depmod: ELF file not for this architecture
| depmod: ELF file not for this architecture
| depmod: ELF file not for this architecture
| depmod: ELF file not for this architecture
| depmod: ELF file not for this architecture
| depmod: ELF file not for this architecture
| depmod: ELF file not for this architecture
| depmod: ELF file not for this architecture
| depmod: ELF file not for this architecture
| depmod: ELF file not for this architecture
| depmod: *** Unresolved symbols in /lib/modules/2.2.17/misc/b1dma.o
| depmod: *** Unresolved symbols in /lib/modules/2.2.17/misc/b1pci.o
| depmod: *** Unresolved symbols in /lib/modules/2.2.17/misc/c4.o
| depmod: *** Unresolved symbols in /lib/modules/2.2.17/misc/cs4231.o
| depmod: *** Unresolved symbols in /lib/modules/2.2.17/misc/eicon.o
| depmod: *** Unresolved symbols in /lib/modules/2.2.17/misc/hisax.o
| depmod: *** Unresolved symbols in /lib/modules/2.2.17/misc/icn.o
| depmod: *** Unresolved symbols in /lib/modules/2.2.17/misc/sc.o
| depmod: *** Unresolved symbols in /lib/modules/2.2.17/misc/t1isa.o
| depmod: *** Unresolved symbols in /lib/modules/2.2.17/misc/t1pci.o
| modprobe: Can't locate module lp
| CSLIP: code copyright 1989 Regents of the University of California
| SLIP: version 0.8.4-NET3.019-NEWTTY-MODULAR (dynamic channels, max=256).
| SLIP linefill/keepalive option.
| PPP: version 2.3.7 (demand dialling)
| PPP line discipline registered.
| PPP BSD Compression module registered
| INIT: Entering runlevel: 3
| Going multiuser...
| neighbour table overflow
| neighbour table overflow
| /sbin/ldconfig: warning: can't open /usr/lib/libltdl.so.0 (No such file or directory), skipping
| /sbin/ldconfig: warning: can't open /usr/lib/libdb1.so (No such file or directory), skipping
| Couldnt get a file descriptor referring to the console
| INIT: Id "c1" respawning too fast: disabled for 5 minutes
| INIT: Id "c2" respawning too fast: disabled for 5 minutes
| INIT: Id "c3" respawning too fast: disabled for 5 minutes
| INIT: Id "c4" respawning too fast: disabled for 5 minutes
| INIT: Id "c5" respawning too fast: disabled for 5 minutes
| INIT: Id "c6" respawning too fast: disabled for 5 minutes
| INIT: no more processes left in this runlevel
Obviously something is wrong with depmod or the modules. Maybe depmod got
compiled for Sparc, but looking for the Intel architecture flags?
There seems to be something wrong with the neighbor routing. Maybe this was
the networking problem you experienced. I'd suggest compiling that feature
out of the kernel for now.
Probably need to do an ldconfig after everything is installed, with the -r
option.
And I'm guessing the installed inittab was one for video. I'm thinking
maybe Slackware needs a configuration setup prompt to ask for whether or not
virtual console logins should be enabled, and likewise for serial ports. It
could be a list with [X] choices much like the prompt for which series to
install, one per tty device to start getty on.
Damn, I *almost* did Ctrl-Alt-Del at the minicom screen on this Intel box
trying to reboot the Sparc 5 :-)
BTW, the SysRQ feature isn't working on Sparc serial port. BREAK just drops
to PROM.. My thought for the kernel is to change it so BREAK doesn't do
that, but make a SysRQ letter to drop to the PROM if SysRQ is enabled.
Another suggestion for the kernel people instead of us.
I boot back to the CD and mounted /dev/sda1 to fix inittab, turning off the
6 vc entries and enabling the 2 serial entries. And the installed
/etc/securetty needs to have the serial ports added, too.
BTW, this is how I'm booting up right now:
ok boot cdrom serial console=ttya root=/dev/sda1 ro
It would be nice if the fsck didn't so that progress bar on serial port. It
produces a LOT of output which slows it down on a 9600 baud serial port.
I'll hack around on that later.
Here's what things look like on my screen right now:
| esp0: target 3 [period 100ns offset 15 10.00MHz FAST SCSI-II]
| SCSI device sda: hdwr sector= 512 bytes. Sectors= 4194995 [2048 MB] [2.0 GB]
| esp0: target 5 [period 100ns offset 15 10.00MHz FAST SCSI-II]
| SCSI device sdb: hdwr sector= 512 bytes. Sectors= 4197405 [2049 MB] [2.0 GB]
| NET4: Ethernet Bridge 007 for NET4.0
| early initialization of device brg0 is deferred
| sunlance.c:v1.12 11/Mar/99 Miguel de Icaza (miguel@nuclecu.unam.mx)
| eth0: LANCE 08:00:20:22:ed:8a
| eth0: using auto-carrier-detection.
| brg0d: network inter for Ethernet Bridge 007/NET4.0
| brg0: generated MAC address FE:FD:0B:54:5A:68
| brg0: attached to bridge instance 0
| Partition check:
| sda: sda1 sda2 sda3
| sdb: sdb1 sdb3 sdb4
| RAMDISK: Compressed image found at block 0
| VFS: Mounted root (ext2 filesystem).
| VFS: Mounted root (ext2 filesystem) readonly.
| change_root: old root has d_count=1
| Trying to unmount old root ... okay
| INIT: version 2.78 booting
| Adding Swap: 50912k swap-space (priority -1)
| /etc/rc.d/rc.S: Testing filesystem status: Read-only file system
| Parallelizing fsck version 1.19 (13-Jul-2000)
| /dev/sda1 was not cleanly unmounted, check forced.
| /dev/sda1: 8257/506880 files (0.1% non-contiguous), 29031/506540 blocks
| Remounting root device with read-write enabled.
| /dev/sda1 on / type ext2 (rw)
| none on /dev/pts type devpts (rw,gid=5,mode=620)
| none on /proc type proc (rw)
| Setting system time from hardware clock...
| Module dependencies up to date (no new kernel modules found).
| modprobe: Can't locate module lp
| CSLIP: code copyright 1989 Regents of the University of California
| SLIP: version 0.8.4-NET3.019-NEWTTY-MODULAR (dynamic channels, max=256).
| SLIP linefill/keepalive option.
| PPP: version 2.3.7 (demand dialling)
| PPP line discipline registered.
| PPP BSD Compression module registered
| Using /etc/random-seed to initialize /dev/urandom...
| INIT: Entering runlevel: 3
| Going multiuser...
| neighbour table overflow
| neighbour table overflow
| /sbin/ldconfig: warning: can't open /usr/lib/libdb1.so (No such file or directory), skipping
| Couldnt get a file descriptor referring to the console
|
|
| Welcome to Linux 2.2.17 (ttyS0)
|
| darkstar login: root
| Password:
| Linux 2.2.17.
| You have mail.
| root@darkstar:~# df
| Filesystem 1k-blocks Used Available Use% Mounted on
| /dev/sda1 1962624 52684 1808632 3% /
| root@darkstar:~#
I can't do much more at this point due to lack of networking commands to get
the ethernet going. So when the full ISO gets synced up, I'll use that.
I've got another approach I'm going to try instead of copying CDs. I just
realized I have an old 2 gig SCSI HD in a Sun 411 case. I can sync the
files onto it to be the install source and just use the CD to boot from.
-- ----------------------------------------------------------------- | Phil Howard - KA9WGN | Dallas | http://linuxhomepage.com/ | | phil-nospam@ipal.net | Texas, USA | http://phil.ipal.org/ | -----------------------------------------------------------------
This archive was generated by hypermail 2b30 : Thu Sep 19 2002 - 11:00:02 PDT