Re: Need help [Fwd: Bug#52468: wml: undefined symbol: PL_sv_undef at /usr/lib/perl5/5.004/sparc-linux/DynaLoader.pm line 169.]
Brian == Brian Bassett [EMAIL PROTECTED] writes: Brian Hello, I got this bug report, and I'm not quite sure Brian whether this is a Perl problem, a Sparc problem, a Perl on Brian Sparc problem, an autobuilder problem, or what. This works Brian for me on an i386. WTF is wml doing building its own perl library? That is Wrong, Evil and Bad (tm). [EMAIL PROTECTED]:logos make wml -q -o UNDEFuEN:index.en.html index.wml Can't load '/usr/lib/wml/perl/lib/5.005/sparc-linux/auto/IO/IO.so' for module IO: /usr/lib/wml/perl/lib/5.005/sparc-linux/auto/IO/IO.so: undefined symbol: PL_sv_undef at /usr/lib/perl5/5.004/sparc-linux/DynaLoader.pm line 169. The problem is that perl as part of wml has been misbuilt somehow. My guess is something is playing games with the namespace cleanup that occured between 5.004 5.005. The correct fix is to use the existing perl environment and link against libperl.a. Of course, the *really* correct fix is for perl5.005 to provide a libperl.so. -- Stephen So if she weighs the same as a duck, she's made of wood And therefore?... A witch!
Re: Netscape for SPARC
On Sun, 12 Dec 1999, Volker Gülke wrote: Hello everybody, I have a Debian SPARC on a SPARCstation 10 installed. Now I'm looking for a good Webbrowser. I hear that there is (or was) a Version of Netscape available for SPARC Linux. Can somebody tell me a place where I can download it or recommend another good browser? Regards, Volker I see, in dselect, a series of netscape*, navigator*, communicator* packages. Alternatively, I remeber pulling down navigator 4.x from the netscape site Sanjeev Ghane Gupta Tel: +91(11) 6941831, 6946619 Eurolink Systems LtdFax: +91(11) 6943732 New Delhi, India email: [EMAIL PROTECTED] Eurolink doesn't pay me to speak for it, so I don't Quid quid latine dictum sit, altum viditur
Re: 'console-tools' cannot find console
On Sun, 12 Dec 1999 [EMAIL PROTECTED] wrote: On Sun, 12 Dec 1999, Ben Collins wrote: On Sun, Dec 12, 1999 at 09:31:05AM -0800, [EMAIL PROTECTED] wrote: On Sun, 12 Dec 1999, Ben Collins wrote: On Sun, Dec 12, 1999 at 08:19:27AM -0800, [EMAIL PROTECTED] wrote: `Couldnt get a file descriptor referring to the console' I get this message every time I boot my Sparc. Now 'console-tools' is erroring when dpkg tries to install it. What sould I look for? crw-r--r--1 root root 5, 1 Dec 11 18:36 /dev/console I have my /dev/console set up for 2.2.x kernels, and it seems to work perfectly fine. IMO, it is a bug in console-tools, but I can't pinpoint the problem. Should file a bug on that package marked important for severity. Would it matter that my console is on ttya? No, because mine isn't, yet I have the same problem. Actually, I think I had the same problem before installing console-tools (which the UTF-8 support needs) I did an xtrace on console-tools' postint, and it looks like the offender is actually debconf in this case. ::++ exec /usr/share/debconf/frontend ./console-tools.postinst configure ::Couldnt get a file descriptor referring to the console Although... I just switched my development lintel to serial console and I sometimes get the same error under other circumstances. I'm still working on that one. -- Ferret no baka
Re: Problems with SILO (Sparc slink)
On Tue, Dec 14, 1999 at 05:43:17PM +0100, Detlev Zundel wrote: Ah - this reminds me - the explanation of hiding the online help for the fdisk `s' command seems completely besides the point to me. Why shouldn't we then hide the man-page for `rm' or other potential dangerous things?? Is the consensus of this list that this is ok or should I file a bug report against fdisk? It'll be fixed in potato. Eric Delaunay is working on it. Ok, I have two remaining problems: The 2.0.34 kernel initialized the text mode correctly as opposed to the 2.2.13 kernel. The difference can be seen in the log messages: 2.0.34 kernel: Dec 9 22:37:23 poc kernel: Console: 16 point font, 864 scans Dec 9 22:37:23 poc kernel: Console: mono SUN 128x54, 1 virtual console (max 63) 2.2.13 kernel: Dec 14 12:55:12 poc kernel: Console: colour dummy device 80x25 How can I get the new kernel to do the same initialization? The difference for me is that I cannot see the leftmost 4 characters on each line... (and the Sun btw supports no color) Which kernel options did you select for the Sparc console fonts support ? The other problem is to get X running - As the IPC uses the BWtwo framebuffer I thought I could fix it by applying the patch from the Ultralinux FAQ to the kernel but the patch is already in there. So I'll better show how X fails: PEXExtensionInit: Couldn't open default PEX font file Roman_M X connection to :0.0 broken (explicit kill or server shutdown). I have the xfonts-pex package installed which owns a file `/usr/X11R6/lib/X11/fonts/PEX/Roman_M.phont'. Unfortunately I cannot strace the X server as strace dies like this: open(/etc/ld.so.preload, O_RDONLY)= -1 ENOENT (No such file or directory) open(/etc/ld.so.cache, O_RDONLY) = 3 open(/lib/libc.so.6, O_RDONLY)= 3 open(/etc/X11/Xserver, O_RDONLY) = 3 syscall: unknown syscall trap 9c23a018 e0002cc0 Are you running slink or potato ? What's the content of /etc/X11/Xserver and could you retry the strace as root ? Greetings, Christian -- Christian Meder, email: [EMAIL PROTECTED] What's the railroad to me ? I never go to see Where it ends. It fills a few hollows, And makes banks for the swallows, It sets the sand a-blowing, And the blackberries a-growing. (Henry David Thoreau)
Re: missing keys in X11
On Tue, Dec 14, 1999 at 09:06:50PM +0100, Volker Gülke wrote: Hello everybody, I have a SUN SPARCstation 10 running with debian slink and problems with missing keys under X11. The dmesg shows me that I am using a SUN Type 5 keyboard without keyclick in a German version. My problem is now, that I'm missing the | symbols which are located on the same key (keynumber 124 - taken from showkey). My questions are now: - where is the correct configuration file for my keyboard? - which are the correct parameters I need to add for the three keys above? - is there somebody who have a complete defined keytable for X and a German SUN Type 5 keyboard who might send it to me? On SparcLinux there is no config file for the Xserver. You just use the commandline arguments. You should try something like: -xkbmap type5_de_nodeadkeys If you look in the file /usr/X11R6/lib/X11/xkb/keymap/sun/de you'll see the different possible maps for de mapping. Greetings, Christian -- Christian Meder, email: [EMAIL PROTECTED] What's the railroad to me ? I never go to see Where it ends. It fills a few hollows, And makes banks for the swallows, It sets the sand a-blowing, And the blackberries a-growing. (Henry David Thoreau)
Re: Problems with SILO (Sparc slink)
On Wed, 15 Dec 1999, Christian Meder wrote: Which kernel options did you select for the Sparc console fonts support ? For the logeintries of my last mail - none really as I used `kernel-image-2.2.13' straight out of the box. By now I have rolled my own kernel with the following options: [*] PROM console [*] Support Frame buffer devices [*] SBUS and UPA framebuffers [ ] CGsix (GX,TurboGX) support [*] BWtwo support [ ] CGthree support [ ] TCX (SS4/SS5 only support) [ ] CGfourteen (SX) support [ ] Leo (ZX) support Virtual Frame Buffer support (ONLY FOR TESTING!) [ ] Advanced low level driver options [*] Support only 8 pixels wide fonts [*] Sparc console 8x16 font [ ] Select other fonts But this changed none of the symptoms - the logentry is still the same. (There's not much help for most options by the way) Are you running slink or potato ? What's the content of /etc/X11/Xserver and could you retry the strace as root ? I am now running potato. /etc/X11/Xserver: - /usr/bin/X11/Xsun Console The first line in this file is the full pathname of the default X server. The second line shows who is allowed to run the X server: RootOnly Console (anyone whose controlling tty is on the console) Anybody - The strace was run as root. If I simply strace without the -e option it gives /lots/ of output and concludes with read(7, , 8192) = 0 read(7, , 8192) = 0 read(7, , 8192) = 0 close(7)= 0 rt_sigprocmask(SIG_BLOCK, [IO], NULL, 8) = 0 rt_sigprocmask(SIG_UNBLOCK, [IO], NULL, 8) = 0 --- SIGSEGV (Segmentation fault) --- +++ killed by SIGSEGV +++ (fd 7 is /etc/X11/banner/Xbanner1.ras) So I guess it is not really connected with the font-file. Thanks for the help Detlev -- 14474011154664524427946373126085988481573677491474835889066354349131199152128 If you know why this number is perfect - you're probably a mathematician...
reboot: PROM prompt loops
Hi there, if I reboot my Sun SLC here with reboot -f it starts rebooting. Then the PROM prompt is displayed in an endless loop: type b (boot), c (continue) or n (new command mode) type b (boot), c (continue) or n (new command mode) type b (boot), c (continue) or n (new command mode) type b (boot), c (continue) or n (new command mode) ... The machine is booted via TFTP and gets all its partitions over NFS. Any clues..??
Re: reboot: PROM prompt loops
On Wed, 15 Dec 1999, Andreas Jaehnigen wrote: Hi there, if I reboot my Sun SLC here with reboot -f it starts rebooting. Then the PROM prompt is displayed in an endless loop: type b (boot), c (continue) or n (new command mode) type b (boot), c (continue) or n (new command mode) type b (boot), c (continue) or n (new command mode) type b (boot), c (continue) or n (new command mode) ... The machine is booted via TFTP and gets all its partitions over NFS. Any clues..?? Had the same problem with my diskless SLC. Easy fix. At the 'ok' prompt type: setenv sunmon-compat? false Apparently the problem is an interaction between the compatibility mode (old) and linux. If you have it in 'new command mode' all the time, you won't have this problem. --Kurt
Help! Kernel panics..
For a couple of weeks I've been trying to track down the mysterious lockups on this Ultra 10 - originally I thought it was due to a weird install (bad first disc), and after reinstalling it seemed fine. Then it went down this weekend, after most of a week of uptime, and perusing bugs.debian.org (for unrelated issues) I saw it might be related to postgres. So I removed postgres, and this time I happened to be logged on when it died: Message from [EMAIL PROTECTED] at Wed Dec 15 16:15:45 1999 ... beaker kernel: Kernel panic: Wheee. Kernel does fpu/atomic unaligned load/store. Unfortunately, the machine has no console (its keyboardless, mouseless and headless, sitting in a rack) so I don't have any more of a message than that. (And I can't build a magic-sysreq kernel, due to strange issues with the 64-bit linker scripts..) And just as a final kick in the pants, its the Big Brother display and paging host, so when it dies it can't even tell me.. :( If I can't get this resolved pretty fast we're going to have to go back to using Solaris on this (and several other) Ultras, so any help is -much- appreciated. -- Shawn McCarthy Systems Engineer 905-A Commerce Rd. Annapolis, MD 21401 410-571-8580 ext. 2241 FAX 410-266-5870 [EMAIL PROTECTED] www.mentorlabs.com
