Re: compiling kernel ...
so ... why this happens to me ??? =( There are some things inthe 3com driver that the compiler doesn't like. It's a known issue, but the fact is, it isn't getting fixed, that I know of. is this related to the 64 bits compiler Yes i've to install egcs64 before i start to compile, because make-kpkg requires sparc64-linux-gcc can i compile a new kernel for the ultra using sparc32 ??? You cannot use a 32bit compiler to create a 64bit kernel. i've tried with 2.2.15 from a deb package and 2.2.16 from ftp.kernel.org None will work. This wont be resolved until gcc3 makes it into woody, and we start using it to compile kernels. -- ---===-=-==-=---==-=-- / Ben Collins -- ...on that fantastic voyage... -- Debian GNU/Linux \ ` [EMAIL PROTECTED] -- [EMAIL PROTECTED] -- [EMAIL PROTECTED] ' `---=--===-=-=-=-===-==---=--=---'
Re: Sparc 20 Quad 100 with 24Bit
* Ben Collins [EMAIL PROTECTED] [001117 20:55]: Might want to try the XF 4.0.1 xserver. It's a huge improvement for my creator card. That would be impressive considering how much better the Creator 3D on my Ultra 1/170E performs over the cg14 on my SS20 2xSM71... but sure, I'll give it a whirl soon. :)
Re: Processes 1-6 Respawning Too Fast
* Ben Roberts [EMAIL PROTECTED] [001118 19:23]: About every seven minutes, the following shows up in /var/log/syslog: Nov 18 15:55:06 starwalker init: Id 2 respawning too fast: disabled for 5 minutes This looks like the gettys might be doing this. Are you using this machine locally? If you are, I'm surprised you get any work done :-) the console should show these same messages, and no login prompt. Those ID's are given in /etc/inittab, by the way. That's where these processes should be listed. If you have not touched this file yet, it's likely there is something wrong with getty. make sure getty really isn't running, then try running 'getty /dev/tty1' and see what happens on the console. There's no reason why you can't spawn a getty for each of the virtual consoles _and_ for the serial port. If you do this and then use the serial console, then you will see these messages on the console as warnings... the system will work fine otherwise. All that needs to be done is to comment out the lines for tty1 through tty6 in /etc/inittab, and that should suffice. You can make this modification and then use the telinit program to reread the inittab file without needing to reboot.
Re: sparcstation 10 w/ X
* Jim Kovalchuk [EMAIL PROTECTED] [001118 19:53]: the only thing that dmesg displays is : fb0: cgsix at e.2000 TEC Rev 2 CPU 68020 Rev 7 [GX] what kinda frame buffer is this ? A cg6 (aka. GX) is an 8-bit framebuffer. There are four variations of the card, but all are 8-bit only. Also, you mentioned trying 16-bit... there are no 16-bit sbus framebuffers as far as I know... and none supported at least. All are either 8-bit or 8 and 24-bit. Unless you have the rare SS10SX (which has an onboard cg14 and you just need to add the VSIMM for it's video memory, etc.), your only 24-bit video option for it would be a Leo/ZX framebuffer... and those cost a pretty penny.
Re: sparcstation 10 w/ X
On Sat, 18 Nov 2000, Jim Kovalchuk wrote: the only thing that dmesg displays is : fb0: cgsix at e.2000 TEC Rev 2 CPU 68020 Rev 7 [GX] what kinda frame buffer is this ? Its an 8 bit frame buffer with hardware accel, use xserver-xsun and she'll run fine. Regards, Peter Firmstone. On Sun, 19 Nov 2000, Q89029292 wrote: On Sat, 18 Nov 2000, Jim Kovalchuk wrote: Hi, just installed potato on my sparcstation 10. Seems i cant run X in 16bpp mode. Is this a hw limitation ? Does my video buffer not have enough memory to run in 16bpp mode ? How do i find out what video buffer i have ? from the command line enter the command $ dmesg the look for the line that begins with fb0: Your card will either be 8 bit or 24 bit color. Regards, Peter Firmstone.
SunOS libraries for emulation
Hi, Does anyone on the list have a copy of the SunOS libraries and files necessary for emulation for SunOS 4.1.x? Thanks in advance, Peter Firmstone.
Re: Help Starting on a SPARCstation2
On Sat, 18 Nov 2000, Sabino Maggi wrote: I've recently picked up the SparcStation 1, and while I make it all the way through the installation and partition the drive as per instructions, I cannot seem to boot into Linux. I reboot, and it gives me the same message it gave me when I booted it and the HD was formatted blank. - No luck at all. Probably you did not create a Sun disk label when partitioning your hd. Nope. Did that. Any other ideas? -- Mike Hoye The only thing you will ever be able to say in the so-called 'social' sciences is: 'some do, some don't.' -Ernst Rutherford
Re: nightmare: apt slink-potato using cds
Thanks to the people who helped me get past the apt-get refusing the cd problem. I had to change the contents of /var/state/apt/cdroms.list (otherwise known as the cdrom index file). Now I find that: 1) the depmod -a in the /etc/init.d gives a _lot_ of symbol not found warnings, because it keeps finding the modules in the /lib/modules/2.2.1 directories (I've upgraded to 2.2.17). AFAICT, the config files are correct (I believe I just accepted all the new ones in the install). modprobe -c does not list any 2.2.1 directories. My 2.2.1 modules are in 2.2.1 directories (not 2.2 directories). I don't have any 2.2 directories in /lib/modules. 2) ipchains has turned off my email. Ie, I have been downloading email from my isp using fetchmail, then sending it to a local MTA (exim) for delivery locally. ipchains by default denies email originating locally (or something: I'm no expert on ipchains or networks. Quite the opposite). I tried turning it off by moving everything related to ip chains out of my rc*.d directories and into rc*.comment directories, and rebooting. Surprise! the ipchains still default to deny. I can get rid of all the rules and set the defaults to ACCEPT with ipchains, but it doesn't persist after reboots. (It's not like I'm on the internet any time other than when I'm downloading mail, and my modem is 14.4K, and the isp assigns me a different isp each time I dial in.) 3) named doesn't seem to resolve even the localhost's name any more. Maybe that's related to the ipchains. $ nslookup seal *** Can't find server name for address 127.0.0.1: No response from server (finds a server at my isp, who not surprisingly doesn't know about my own hosts name (i'm in the middle of downloading mail as I write this)) $ ps ax | grep named (doesn't find any named process - and I didn't kill any either) And there _is_ an /etc/rc2.d entry of S19bind I just don't get it. 4) xdm won't let me log in (neither root nor me) I imagine this problem will go away when I get the above solved. I've moved xdm out of the rc*.d directories too. 5) gpm keeps starting up and writing piles of junk in my logs. I keep killing it off and it keeps coming back. At least this one went away when I took out the gpm stuff from my init directories. 6) while I was in the init directories I saw stuff related to tamagotchi. What is tamagotchi, tamad, tama? It doesn't exist in any of my man files (but that's not surprising, it's surprisingly difficult to get _all_ the man pages installed) nor in `ls -lad /usr/doc/tama*`. H! If I'm lucky, I'll be able to hack this machine so I'll be able to read your answers later. Sigh. -- [EMAIL PROTECTED] -- Hi! I'm a .signature virus! Copy me into your ~/.signature, please! --
Re: nightmare: apt slink-potato using cds
Allright, I've gotten rid of ip* from the cron.* directories.
So presumably firewall stuff is not why named is failing
(unless the current version depends on it? who knows)
On Sun, Nov 19, 2000 at 12:25:18PM -0500, Brenda J. Butler wrote:
3) named doesn't seem to resolve even the localhost's
name any more. Maybe that's related to the ipchains.
$ nslookup seal
*** Can't find server name for address 127.0.0.1: No
response from server
(finds a server at my isp, who not surprisingly doesn't
know about my own hosts name (i'm in the middle of
downloading mail as I write this))
$ ps ax | grep named
(doesn't find any named process - and I didn't kill any
either)
And there _is_ an /etc/rc2.d entry of S19bind
I just don't get it.
Ok, named was failing to start because I had moved /etc/bind
to /etc/bind.comment. Why does named use bind conf files,
when it has its own conf files (with largely the same info)
in /etc/named.conf and /var/named?
And now that named _is_ running (I've moved the /etc/bind.comment
back to /etc/bind), it still fails to resolve even the local host
name (in this example, I'm _not_ dialed in to my isp):
# nslookup seal
Server: localhost
Address: 127.0.0.1
*** localhost can't find seal: No response from server
# ps ax | grep named
142 ? S 0:00 /usr/sbin/named
other named entries such as the grep, man, and info
#
=== contents of resolv.conf: ==
nameserver 127.0.0.1
search stuffedanimals
nameserver 209.151.0.10
nameserver 209.151.0.12
=== contents of /etc/named.conf: ==
// generated by named-bootconf.pl
options {
directory /var/named;
/*
* If there is a firewall between you and nameservers you want
* to talk to, you might need to uncomment the query-source
* directive below. Previous versions of BIND always asked
* questions using port 53, but BIND 8.1 uses an unprivileged
* port by default.
*/
// query-source address * port 53;
};
//
// Boot file for name server
//
// type domain source file
zone . {
type hint;
file named.root;
};
// Zone boot information and daemon options are kept in other files
// (autoincluded from boot.zones)
//
// Name server zone boot file
// See named(8) for syntax and further information
//
// type domain source file
// (autoincluded from boot.options)
//
// Options for name server
// Use `bindconfig' to automatically configure this file
//
// type domain source file
zone localhost {
type master;
file named.local;
};
zone 127.in-addr.arpa {
type master;
file named.rev-local;
};
// Custom configurations below (will be preserved)
=== contents of /etc/host.conf: ==
order hosts,bind
multi on
=== contents of /etc/hosts: ==
127.0.0.1 localhost
192.168.110.5 seal.stuffedanimals seal # sun sparc station 20
# ...some other machines, all on 192.168.110
# The following lines are desirable for IPv6 capable hosts
# (added automatically by netbase upgrade)
::1 ip6-localhost ip6-loopback
fe00::0 ip6-localnet
ff00::0 ip6-mcastprefix
ff02::1 ip6-allnodes
ff02::2 ip6-allrouters
ff02::3 ip6-allhosts
===
And why won't named dump its cache to /var/tmp/named_dump.db
when I send it the INT signal, as the man page says it should?
Where is the _PATH_DUMPFILE variable supposed to be defined?
In the environment? I don't have any _PATH_DUMPFILE in
my environment. And when I put one there and signal the
named, it still doesn't work:
# export _PATH_DUMPFILE=/var/tmp
# kill -HUP 142
# ls /var/tmp
vi.recover
4) xdm won't let me log in (neither root nor me)
I imagine this problem will go away when I get the above
Hmm, upon reflection, perhaps xdm still won't work. It
was complaining about permission problems in /var/log/xdm.log,
and I suppose it wouldn't be logging those if ipchains was
preventing it from receiving anything.
argh. Well, first to fix the named problem, then the
xdm after. To fix the xdm problem, I suppose I'll have
to learn all about xauth/xdm interaction.
--
[EMAIL PROTECTED]
--
Hi! I'm a .signature virus! Copy me into your ~/.signature, please!
--
Re: compiling kernel ... and the trashing system
well ... finally i compiled a new kernel (2.2.16) ommiting the 3com and a few other drivers that were causing me problems now my system is working ok and i haven't see it trashing again You cannot use a 32bit compiler to create a 64bit kernel. i was sure it wasn't a good idea =) thanks a lot to everyone on the list esteban
Potato revision 2
Hello world, In the next day or so, Ben Collins [EMAIL PROTECTED] will be creating a list of show stoppers for potato revision two. This list will hopefully be maintained for future revisions, and be treated more or less as the release critical bug list has been during the freeze [0]. If there's anything that needs to be done, or that's in the process of being done, that we should be waiting for for r2 rather than leaving until r3 or later, please make sure Ben knows about it. r2 will happen ASAP after that list is emptied. The more detailed the information you can send the better: we need new boot-floppies is better than no! we can't release yet, and I'm preparing new boot-floppies for i386 to fix bugs 32343 55643 and 76432, which will be ready in a day; updates for powerpc are also needed (bug 65653) and Daniel Jacobowitz is taking care of that, but I don't know how long it will take; updates for other architectures aren't urgent is better still. Who's doing it, how long it will take, why it's needed, and why it won't cause new problems would be ideal. For reference, you can consider the list to currently be: Show stoppers for potato r2 ~~~ * Show stopper list needs to be created (Ben Collins) (will be done by around (2000/11/21 12:00 UTC) * 2.2.18 boot-floppies for i386 (Adam di Carlo) (will be done by around (2000/11/22 12:00 UTC) Note further that r2 is expected to happen fairly soon now (days, not weeks) [1]. Cheers, aj [0] ie, we'll try to make sure there aren't any left; but if there are some that don't appear to have any hope of ever being fixed, they'll eventually just get ingored. [1] cf http://lists.debian.org/debian-announce-00/msg00011.html -- Anthony Towns [EMAIL PROTECTED] http://azure.humbug.org.au/~aj/ I don't speak for anyone save myself. GPG signed mail preferred. ``We reject: kings, presidents, and voting. We believe in: rough consensus and working code.'' -- Dave Clark pgpx1mRZ9WSLl.pgp Description: PGP signature
