Re: booting ultra 1
On 06/19/06 01:14:13AM -0400, Matthew Pease wrote: Oh, man, what havent I tried :) I guess this is not as simple as a command to tell the thing where to find the next piece of code it needs to boot? Surely there is someone with an Ultra 1 that has run into this! The iso just posted to this list gave some different startup messages, and looked promising, but then barfed like the rest. I used to have an Ultra1 running sarge, but had to retire it because of hardware problems. Please post exactly what happens when you just hit enter at the silo prompt. If you don't have a serial console to copy/paste the output feel free to take a picture and post a URL containing the error messages. Have you tried net booting it? Setting up a RARP and TFTP isn't that difficult and would rule out a broken CDROM. Am I putting myself through needless pain? Is it just uncool to use anything but an i386? It should work fine, I had virtually no problems setting up Debian on my Ultra1, although I believe I installed woody and dist-upgraded to sarge. Jim. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: tg3 driver fails
On 07/06/05 10:22:06PM +0200, Hendrik Sattler wrote: Am Mittwoch, 6. Juli 2005 22:11 schrieb David S. Miller: However, I agree that all firmware should be removed from the kernel and provided externally (like done with the AVM B1 ISDN cards). However, Debian _must_ tell then, where to find the firmware (means: don't let the user alone with this)! This doesn't work either. What if the tg3 card is your only network access? How might you get that magic firmware file? Does initrd or the like work with network boot? How does the card actually do network boot when not being able to operate without firmware (chicken-egg problem)? If the network card cannot handle network boot: why would you want such a crap? This scheme doesn't work, firmware belongs in the kernel tree so that at boot time, regardless of your internet access, you can get a working device. I could give you various reasons why firmware does _not_ belong into a compiled driver. A clean interface in driver development and easy firmware replacement are just two things. I think he meant packaged with the kernel and not built statically into the kernel binary; although if the firmware is sufficiently separate and the build process just links it in I don't see why that would be a problem either, sort of like how nVidia does their driver. Even barring that, you could easily throw it in an initrd or initramfs image to make it available for early initialization. What if I want to use NFS root over the tg3 device? See above. I find it strange that there are network cards that need a firmware uploaded :-/ It's good for ISDN cards as there are many modes to operate in but for an ethernet card? HS Jim. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Question on apt-settings
On 06/17/05 07:31:31AM +0200, Jurzitza, Dieter wrote: Hi folks, since I haven't experienced a crossing over from a testing to a stable distribution: is it now the time to change the entries in /etc/apt/sources.list from testing to stable, as I do not plan to further proceed from sarge to sid for the moment? Especially because within sources.list there is no reference to the distribution's name! Many thanks for your inputs, take care The time to do that was at release, but if you haven't updated anything since then it won't hurt to do it now. And if you leave it set to testing you'll upgrade to etch, not sid. sid will always be unstable and will never become testing. Dieter Jurzitza Jim. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Low latency sparc64 kernels ?
On 06/13/05 11:55:55PM +0100, Martin wrote: On Sat, 2005-06-11 at 14:36 -0700, [EMAIL PROTECTED] wrote: Premeptibility patching is broken on Sparc. ReiserFS is broken on Sparc. Broken as in doesn't work at all, doesn't work fast or occasionally causes major corruption and eats all of your data. I have a sarge machine running / on reiserfs and it seems to work - can you provide any more pointers on this as I was just about to move some important data on to it. A while ago, not sure how many months it was, I had a reiserfs partition get partially corrupted from an unclean shutdown and the only evidence was that the box would oops and blank the screen when syslogd started. I had to hook up a serial console to even see the oops, I assume reiserfs was overwriting something in memory causing the display corruption. Mounting it said it was clean, reiserfsck said it was clean and yet it still hung when syslogd started. F'd with it a little more, figured out which file it was, deleted it, tar'd up /var and moved it to XFS and haven't had a problem since. Every time I figure it must have matured and give reiserfs a try it bites me in the ass. IMO XFS is much more mature and the userland tools are much, much better. Cheers, - Martin Jim. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: question about kernel
On 06/07/05 02:01:37PM +0200, gaspo wrote: i have a debian [14:00:[EMAIL PROTECTED]:/home/gaspo]$ uname -a Linux sparc64 2.4.18 #2 Thu Apr 11 14:37:17 EDT 2002 sparc64 unknownwith kernel 2.4.18 basic... i want to upgrade with another new kernel... what is the best way for uypgrade kernel? i can download witrh apt-cache search kernel NEw and dpkg -i KERNEL.deb... work? You can use apt-cache search kernel-image to get a list of the available kernels then just 'apt-get install kernel-image-version' to install it. Then depending on how you have your bootloader setup you might have to update /etc/silo.conf. Jim. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: question about kernel
On 06/08/05 12:41:19AM +0200, gaspo wrote: but when i try apt-get install kernel-image-2.4.19-sun4u Setting up kernel-image-2.4.19-sun4u (26) ... after some SEcond i have the message: depmod: *** Unresolved symbols in /lib/modules/2.4.19/kernel/net/ipv6/netfilter/ip6table_filter.o depmod: *** Unresolved symbols in /lib/modules/2.4.19/kernel/net/ipv6/netfilter/ip6table_mangle.o depmod: *** Unresolved symbols in /lib/modules/2.4.19/kernel/net/ipx/ipx.o depmod: *** Unresolved symbols in /lib/modules/2.4.19/kernel/net/netlink/netlink_dev.o There was a problem running depmod. This may be benign, (You may have versioned symbol names, for instance). Or this could be an error. In any case, since depmod is run at install time, we could just defer running depmod Would you like to abort now? [Yes] Yes dpkg: error processing kernel-image-2.4.19-sun4u (--configure): subprocess post-installation script returned error exit status 1 Errors were encountered while processing: kernel-image-2.4.19-sun4u E: Sub-process /usr/bin/dpkg returned an error code (1) I haven't used a woody system in a while so I don't remember seeing that, but I really doubt you need those particular modules and the rest of the kernel should work just fine so don't abort the install. And as long as you keep your current kernel installed, you'll be able to fall back to it if 2.4.19 does cause you problems. 2005/6/7, Jim Crilly [EMAIL PROTECTED]: On 06/07/05 02:01:37PM +0200, gaspo wrote: i have a debian [14:00:[EMAIL PROTECTED]:/home/gaspo]$ uname -a Linux sparc64 2.4.18 #2 Thu Apr 11 14:37:17 EDT 2002 sparc64 unknownwith kernel 2.4.18 basic... i want to upgrade with another new kernel... what is the best way for uypgrade kernel? i can download witrh apt-cache search kernel NEw and dpkg -i KERNEL.deb... work? You can use apt-cache search kernel-image to get a list of the available kernels then just 'apt-get install kernel-image-version' to install it. Then depending on how you have your bootloader setup you might have to update /etc/silo.conf. Jim. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Bug #251149: gcc wrapper for sparc is chronically broken
On 05/24/05 05:38:58PM -0700, David S.Miller wrote: From: Jim Crilly [EMAIL PROTECTED] Date: Tue, 24 May 2005 14:42:27 -0400 True, but building kernels on sparc64 wasn't terribly fun for me the last time I tried it either so I decided it wasn't worth it and just stuck with the Debian kernel images. Amusing as I do all of the sparc64 kernel development under Debian/Sparc on all of my machines. It builds out of the box, both 2.4.x and 2.6.x. Feel free to laugh at me all you wish. It was quite a few months ago so I don't remember the issues I ran into and for all I know they don't exist any more, I do remember that building the kernel itself was fine but I think I had problems with the userspace stuff like menuconfig, on sparc64 btw. Jim. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Memory not aligned kernel problem
Ben Collins wrote: On Fri, Dec 19, 2003 at 01:15:52PM +0100, Pieter-Paul Spiertz wrote: Hi Marco, On Fri, 19 Dec 2003, marco wrote: I just compiled the 2.6.0 kernel. The compile process is just fin, byt the image i get seems not to be bootable. Right after i select the image at the sil prompt at boot i get a Memory not aligned error and i am kicked to the openboot prompt again. What compiler did you use? I wouldn't trust all GCC versions. GCC 3.2.3 is known to work with the 2.4 kernels. I consider 2.6.0 a 'most things work' milestone. Sparc support for 2.6 hasn't been tested or reported on this list very much lately. It might be broken. In that case, it will be fixed in a next version of the 2.6 series. I'm guessig this is sparc32. I know for a fact that 2.6.0 is pretty solid on sparc64. I just booted 2.6.0 on an Ultra1 and it oopsed in the zilog interrupt handler as soon as the sysv scripts were done when init tries to start getty for the serial console. But since I don't currently have a serial console hooked up I didn't copy the oops, sorry. It happens consistantly on boot though so if you want I should be able to get the oops for you. Thanks, Jim.
Performance Question
I recently setup Courier-IMAP on a Ultra2 with 2x300Mhz CPUs because the old slower hardware and wu-imapd/mbox setup was just getting too slow. For the most part the combination of newer hardware and Maildir was a big difference, especially on bigger mailboxes. The problem now is that on the really large mailboxes (the cut off seems to be around 9K messages) the sort capability causes problems with IMP because it takes so long. If I disable server-side sorting it's magnitudes faster, but IMP loses some functionality. I don't think there's much I can do since it just takes a long time for imapd to sort 9K messages, but I thought I'd ask anyway incase I'm missing something. Recompiling the package wouldn't help anything, right? Would a different imap daemon do better? Thanks for any input, Jim - This mail sent through IMP: http://horde.org/imp/
Re: Custom tftpboot images
I would be interested in knowing how to do this myself, if you could post a link to the docs you have that would be great. Jim. On Wed, 2003-04-02 at 03:53, Antonio Prioglio wrote: On 1 Apr 2003, Irvin Probst wrote: My question is: is there anybody interested in such a tftp boot file for sparc with XFS support ? If I'm the only one I'll make a quick and dirty hack to Ben Collin's tftpboot.img (add mkfs.xfs and piggyback a new kernel), but if other people are interested I'll try to make something cleaner. Hi, I have two U2 that would welcome a move to xfs. If you decide to post something I'll be glad to learn about it. Saluti, Antonio Prioglio [EMAIL PROTECTED] -- /\ \ /ASCII RIBBON CAMPAIGN - AGAINST HTML MAIL X - AGAINST MS ATTACHMENTS / \ -- Help protect your rights on-line. Join the Electronic Frontiers Foundation today: http://www.eff.org/join --- It's funny how the cross - an execution tool - has become a religion symbol.I have to wonder if we'd be seeing Jesus with a stake up his sphincter if that indeed was the was he was executed. I guess if he'd lived in more modern times we'd see Jesus in an electric chair above the alter. --- In-Warren R says: Our technicians are working on it and it will be blazing fast. (ATT Tech support) signature.asc Description: This is a digitally signed message part
Re: mysql-server on woody : bdb problems
I noticed this too, but since there already was a bug filed on Jun21st (#101783) with acknowledgment I didn't say anything. I did get mysqld to start with the --skip-bdb switch and it seems to be fine, although I'm not sure what that might/will effect =) Jim On Sunday 08 July 2001 05:50, Olivier Bornet wrote: Hello, I want to install mysql-server on a GNU/Debian SPARC box running woody. When starting the daemon, it say: 010708 11:25:11 mysqld started 010708 11:25:11 bdb: architecture lacks fast mutexes: applications cannot be threaded 010708 11:25:11 Can't init databases 010708 11:25:11 mysqld ended The version is : ii mysql-server 3.23.36-2 mysql database server binaries I have see the same message in the archive of this list, but I have found no answer. Any hints ? Thanks in advance, and good day. Olivier
Re: Reiserfs
I have no idea if reiserfs runs on sparc or not, I know they're just now getting the Alpha port in working order. The error you're getting is related to a define that they have gotten into the errno.h for i386 but no other arches yet, at the bottom of include/asm/errno.h add an entry like: #define EHASHCOLLISION127 /* reiserfs hash collision */ and that should fix that error. Eventually when reiserfs is in the kernel (around 2.4.1) this should be taken care of. Jim I tried to compile a 2.2.18 with reiserfs-support. This happens: make[3]: Entering directory `/usr/src/linux-2.4.0/fs/reiserfs' gcc -D__KERNEL__ -I/usr/src/linux/include -Wall -Wstrict-prototypes -O2 \ -fomit-frame-pointer -fno-strict-aliasing -m32 -pipe -mno-fpu \ -fcall-used-g5 -fcall-used-g7-c -o namei.o namei.c In file included from namei.c:8: /usr/src/linux/include/linux/reiserfs_fs.h: In function `find_next_nonzero_bit': /usr/src/linux/include/linux/reiserfs_fs.h:2043: warning: implicit declaration of function `generic_ffs' namei.c: In function `reiserfs_add_entry': namei.c:474: `EHASHCOLLISION' undeclared (first use in this function) namei.c: In function `reiserfs_mkdir': namei.c:646: warning: comparison is always false due to limited range of data \ type Well, are there any reports whether reiserfs runs on Sparc/32bit at all? Does it normally compile, or am I doing something wrong (wrong header-file somewhere)? Peter -- Any good Unix security engineer can clean up any Unix box. But I'm not sure there are people even within Microsoft who know how to clean up an NT box. -- Michael Zbouray -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
