Re: Partitioning schemes for partman-auto
Anton Zinoviev [EMAIL PROTECTED] writes: On 1.III.2004 at 20:10 Falk Hueffner wrote: Swap nowadays only serve to spill pages that haven't been accessed for hours, or to slow down the system enough so you can kill processes gone haywire; therefore, it is not performance critical and a swap file will do just fine and is more flexible WRT resizing. I can remember the installer of Red Hat saying that the swap must be always at least as much as the available RAM even if there is enough RAM. Without swap the performance of the system would be degraded. Do you know if this is true? I dimly recall there used to be a bug in older kernel releases to this effect. I don't believe it is relevant any more. If it was, I'd certainly consider it a kernel bug. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Partitioning schemes for partman-auto
Anton Zinoviev [EMAIL PROTECTED] writes: The partitioning schemes of partman-auto currently are only examples. So if you have some idea for usefull partitioning just tell it. I have to know what partitions your scheme has and an example size of each partition as well as the minimal and the maximal usefull size. For example: This is the only partitioning scheme which in my opinion fits for users who might go with whatever the default is: / Whole harddisk swapnone, but 512M swap file (independent of memory) Rationale: I don't see any noticeable benefit for Jane User in having lots of small partitions; and as has been pointed out it always either wastes space or is prone to overruns. Swap nowadays only serve to spill pages that haven't been accessed for hours, or to slow down the system enough so you can kill processes gone haywire; therefore, it is not performance critical and a swap file will do just fine and is more flexible WRT resizing. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Partitioning schemes for partman-auto
Joey Hess [EMAIL PROTECTED] writes: Falk Hueffner wrote: This is the only partitioning scheme which in my opinion fits for users who might go with whatever the default is: / Whole harddisk swapnone, but 512M swap file (independent of memory) I also wouldn't mind having this scheme available, although I think the current simple scheme (/, /home, /swap) is also useful. Fine as long as there is some help text which makes the choice trivially obvious. That seems hard. Swap nowadays only serve to spill pages that haven't been accessed for hours, or to slow down the system enough so you can kill processes gone haywire; therefore, it is not performance critical and a swap file will do just fine and is more flexible WRT resizing. Not true for many of our target architectures and systems. Those are used by people who do this stuff manually anyway. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [Q] make demo fails
[EMAIL PROTECTED] (Erich Waelde) writes: else depmod -q -a -b ./tmp/demo/tree/ 2.4.22-1-386; fi ; == depmod: QM_MODULES: Function not implemented make[1]: *** [demo-tree-stamp] Error 1 make: *** [demo] Error 2 What is the build system trying to tell me? I'm running on kernel 2.6-test9-1 and required tools, is that a problem? Yes, that is most likely the cause. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: /etc/mailname
[EMAIL PROTECTED] writes: A few days ago I installed a Debian system (V3.0r1). When emacs is invoked it says No /etc/mailname. Reverting to default... and waits for 3 seconds. Of course this is very undesirable. Something is broken. Yes, and it's emacs. Waiting 3 seconds is just silly. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: status of pppoe support in d-i ?
Marco d'Itri [EMAIL PROTECTED] writes: [EMAIL PROTECTED] wrote: I tried to rebuild d-i with ppp-udeb.deb in the local-udeb but it do not build anymore: 461 symbols, 375 unresolved Traceback (most recent call last): File /usr/bin/mklibs, line 469, in ? raise No library provides non-weak + symbol No library provides non-weak devnam devnam is a global variable defined in pppd/options.c, so I don't understand why mklibs is complaining. Please post the relevant line of the output of readelf -s -W on the file containing the symol. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Language choser is also keyboard choser ??
Daniele Nicolodi [EMAIL PROTECTED] writes: I dont' like also the way to select language, is not clear. I like more a solution like: Please chose your languge: DE - (I don't know german) EN - Select that entry to procede in English ES - (I don't know espagnol also) IT - Seleziona questa voce del menu per procedere in italiano Whe can order coutry codes alfabetically so, i think, found the right one is more easy. I don't like these lengthy selection texts; they don't really add any information, but make it harder to skim. What's wrong with simply: Deutsch English Espagnol Italiano ? Much clearer, and you don't even have to know your country code. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [Fwd: Alpha Debian Netinst]
From: Andrew Forgue [EMAIL PROTECTED] Well, I'm trying the sarge-alpha-netinst.iso from testing (both daily and non-daily) on an alpha PWS 500au with 128mb ram using aboot/SRM and so far am unsuccessful. It dies while trying to mount /dev/rd/0. If I tell it to use /dev/ram0 as a root, it says it can't find init. If I pass init=/linurc to the kernel, it still says no init found. Try root=/dev/ram initrd=boot/root.bin ramdisk_size=16384 devfs=mount,dall -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Alpha status
Goswin von Brederlow [EMAIL PROTECTED] writes: Also busybox got miscompiled (tr support missing) due to linux-kernel-header bugs (fixed?). Fixed in glibc (#212101), busybox probably needs recompilation, though. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#219928: debian-installer: build failures on sparc
[EMAIL PROTECTED] (Geert Stappers) writes: On Tue, Nov 11, 2003 at 12:02:34AM +0100, Falk Hueffner wrote: Can I get more information on why this is a bug in mklibs? Does some library actually provide timeout_yet? If so, can I please see readelf -s -W output of it? See bug 211092 ( http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=211092 ) It was original titled 'building debian-installer on sparc fails' I cannot find this information there, either. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#219928: debian-installer: build failures on sparc
[EMAIL PROTECTED] (Geert Stappers) writes: On Tue, Nov 11, 2003 at 10:02:44AM +0100, Falk Hueffner wrote: | Can I get more information on why this is a bug in mklibs? See bug 211092 ( http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=211092 ) | Does some library actually provide timeout_yet? | If so, can I please see readelf -s -W output of it? I don't care about a single accidentically missing symbol. I don't understand. Are you saying if a symbol is missing, mklibs should ignore that and create a binary that cannot be executed? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#219928: debian-installer: build failures on sparc
[EMAIL PROTECTED] (Geert Stappers) writes: On Tue, Nov 11, 2003 at 10:54:31AM +0100, Falk Hueffner wrote: [EMAIL PROTECTED] (Geert Stappers) writes: On Tue, Nov 11, 2003 at 10:02:44AM +0100, Falk Hueffner wrote: | | | I cannot find this information there, either. | Can I get more information on why this is a bug in mklibs? See bug 211092 ( http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=211092 ) | Does some library actually provide timeout_yet? | If so, can I please see readelf -s -W output of it? I don't care about a single accidentically missing symbol. I don't understand. Are you saying if a symbol is missing, mklibs should ignore that and create a binary that cannot be executed? I'm only saying that it is a mklibs bug. Where the timeout_yet or whatever noweak symbol is, is not important(to me). Then how do you know it is a bug in mklibs and it is not actually a missing symbol? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#219928: debian-installer: build failures on sparc
Can I get more information on why this is a bug in mklibs? Does some library actually provide timeout_yet? If so, can I please see readelf -s -W output of it? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#218632: srm_env module required by partitioner on Alpha
Package: linux-kernel-di Severity: normal Hi, alpha.sh from partitioner relies on the srm_env module. Please add it to some core modules udeb. Falk -- System Information: Debian Release: testing/unstable Architecture: alpha Kernel: Linux juist 2.6.0-test3 #2 Sat Aug 9 20:21:49 CEST 2003 alpha Locale: LANG=C, [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#218632: srm_env module required by partitioner on Alpha
Steve Langasek [EMAIL PROTECTED] writes: On Sun, Nov 02, 2003 at 01:54:02PM +0100, Goswin von Brederlow wrote: Thorsten Sauter [EMAIL PROTECTED] writes: | alpha.sh from partitioner relies on the srm_env module. Please add it | to some core modules udeb. no idea about alpha machines, but what about the comment line in alpa.sh? # Load srm_env.o if we can; this should fail on ARC-based systems. (modprobe srm_env || true) 2 /dev/null Thats why the possibly failure to load the module is hidden. All non srm based systems should fail to load this and the script shouldn't stop. This part should actually be moved into arch-detect I think. Is it needed for anything else? You mean srm_env? Yes, it's used to get the system language setting by srm-reader. Can partitioner depend on arch-detect being run first? Can you depend on the user not having manually unloaded the module from the second console after arch-detect runs ;P, or will arch-detect be able to pass this information to partitioner in some other way? I have no clue. It would be sensible if this information would be generally available somewhere. We could also detect it by grepping for system serial number.*MILO in /proc/cpuinfo, though. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Ultimate bug: bringing sanity to linux
Dominic-Luc Webb [EMAIL PROTECTED] writes: Those CD images are useless to a newby without the bootup/install diskettes for people who cannot go the network install route. Could you elaborate on that point? Most newbies will probably have PCs bought after 1995, which can boot from CDs. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#215448: di_mem_chunk_alloc() should return sizeof(long) or better alignment
Richard Hirst [EMAIL PROTECTED] writes: On Sun, Oct 12, 2003 at 11:48:07PM +0200, Falk Hueffner wrote: Just use long mem[] there, we can assume a C99 compiler, and it will ensure alignment (at least for all Linux platforms). Really? I thought the code later referenced something like foo-mem[bar] which will surely break if you just s/char/long/ Dunno, might need to add casts there. I would've thought it only casts the value to void*. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#215448: di_mem_chunk_alloc() should return sizeof(long) or better alignment
First, IMHO we should drop this stuff and just use malloc. I don't see
any point in increasing code size and introducing new bugs thereby.
Anyway:
struct di_mem_area
{
di_mem_area *next; /** the next mem area */
di_mem_area *prev; /** the previous mem area */
di_ksize_t index;/** the current index into the mem array */
di_ksize_t free; /** the number of free bytes in this mem area */
di_ksize_t allocated;/** the number of atoms allocated from this area */
char mem[MEM_AREA_SIZE]; /** the mem array from which atoms get allocated
* the actual size of this array is determined by
* the mem chunk area_size. ANSI says that it
* must be declared to be the maximum size it
* can possibly be (even though the actual size
* may be less).
*/
};
I also wonder about MEM_AREA_SIZE being hardwired at 4, although I
didn't dig in to the code far enough to see if it should be 8 on 64
bit platforms.
Just use long mem[] there, we can assume a C99 compiler, and it will
ensure alignment (at least for all Linux platforms).
--
Falk
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#178511: mklibs: Misses some symbols?
Hi, the symbol parsing part of mklibs has been reworked in 0.1.12. Can you please check whether the problem is still there and if so, how exactly to reproduce it? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Milo
Greg Folkert [EMAIL PROTECTED] writes: Okay, in Joey's summary of Oldenberg he hints @ Milo Source being unavailable. Well, there's a version from Stefan Reinauer from Suse (http://www.suse.de/~stepan/), but it is difficult to compile with a non-Suse kernel source. He even has some source based on the 2.4 kernel at ftp.suse.com//pub/people/stepan/milo/new, but I don't know whether it ever worked. Reportedly, the Gentoo folks got a reasonable Milo version to compile recently, so anybody interested in working on it should probably ask there. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: New d-i debcamp in a few months?
How many are interested in going to such event? Where are you located in the world? Me too, but only if it's close and I don't have anything better to do at the time. Count me as approximately 0.4 persons. - Falk Hueffner - Tuebingen, Germany -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#172828: mklibs : problem on mips
Alastair McKinstry [EMAIL PROTECTED] writes: I'm writing to you as the authors of mklibs Damn. Why does everybody think I know how mklibs works just because I wrote it? ;) Seriously, I just stole most stuff from existing scripts, I'm not really an expert on dynamic linking... They are found in libc. Now, in i386, for example, they are of type: eg. NOTYPE WEAK DEFAULT UND _pthread_cleanup_pop_restore But on MIPs, they are all: [EMAIL PROTECTED]:/lib$ readelf -s -W libc.so.6 | grep WEAK | grep UND | grep FUNC 2023: 0 FUNCWEAK DEFAULT UND _pthread_cleanup_push_defer I believe this means they are all weak objects, not normally marked as such. They will resolve to zero if not found elsewhere, but undefined_symbols() in mklibs treats them simply as undefined. I am not quite sure about how to handle weak symbols properly. If they are provided somewhere, we would usually want them to be used. So just ignoring all weak symbols doesn't seem proper. The best way seems to be to check whether the set of unresolved symbols didn't change since the last pass, and if so, whether it consists of only weak symbols. If so, just leave them undefined and quit. Otherwise abort (better than looping infinitely). Does that sound sensible? I'm a bit short on time currently, I'll see whether I can make a patch the next few days. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: d-i: depmod fails on Alpha
Herbert Xu [EMAIL PROTECTED] writes: Falk Hueffner [EMAIL PROTECTED] wrote: I'm trying to build d-i on Alpha. I get this error: # Set up modules.dep, ensure there is at least one standard dir (kernel # in this case), so depmod will use its prune list for archs with no # modules. mkdir -p ./tmp/net/tree/lib/modules/2.4.20-generic/kernel depmod -q -a -b ./tmp/net/tree/ 2.4.20-generic depmod: QM_MODULES: Function not implemented make: *** [net-tree-stamp] Error 1 I'm running kernel 2.5.59. Is that a problem? Or any other ideas? It's running the depmod from module-init-tools. You need to check the existence of module-init-tools and always run the depmod from modutils for a 2.4 kernel. Actually, I don't have module-init-tools installed, since my kernel doesn't have modules support... and I guess that's the problem. You're right that this should be checked, though. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
d-i: depmod fails on Alpha
Hi, I'm trying to build d-i on Alpha. I get this error: # Set up modules.dep, ensure there is at least one standard dir (kernel # in this case), so depmod will use its prune list for archs with no # modules. mkdir -p ./tmp/net/tree/lib/modules/2.4.20-generic/kernel depmod -q -a -b ./tmp/net/tree/ 2.4.20-generic depmod: QM_MODULES: Function not implemented make: *** [net-tree-stamp] Error 1 I'm running kernel 2.5.59. Is that a problem? Or any other ideas? modutils is 2.4.21-2. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#184908: Missing build dependency on gtk
Package: cdebconf Version: unavailable; reported 2003-03-15 Severity: important Justification: fails to build from source Current CVS: [...] Compiling /home/falk/debian/debian-installer/tools/cdebconf/src/modules/frontend /gtk/gtk.c to gtk.opic Package gtk+-2.0 was not found in the pkg-config search path. Perhaps you should add the directory containing `gtk+-2.0.pc' to the PKG_CONFIG_PATH environment variable No package 'gtk+-2.0' found /home/falk/debian/debian-installer/tools/cdebconf/src/modules/frontend/gtk/gtk.c:64:21: gtk/gtk.h: No such file or directory /home/falk/debian/debian-installer/tools/cdebconf/src/modules/frontend/gtk/gtk.c:69: syntax error before GtkWidget /home/falk/debian/debian-installer/tools/cdebconf/src/modules/frontend/gtk/gtk.c:69: warning: no semicolon at end of struct or union /home/falk/debian/debian-installer/tools/cdebconf/src/modules/frontend/gtk/gtk.c:80: syntax error before '*' token [...] Apparently, a build dependency on some gtk devel package is missing. -- System Information: Debian Release: testing/unstable Architecture: alpha Kernel: Linux juist 2.5.59 #4 Sat Jan 18 12:46:41 CET 2003 alpha Locale: LANG=C, [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: working on the Alpha port
Herbert Xu [EMAIL PROTECTED] writes: On Thu, Dec 12, 2002 at 05:12:06PM +0300, Alexander Kotelnikov wrote: On Thu, 12 Dec 2002 11:24:25 +0100 ML == Mario Lang [EMAIL PROTECTED] wrote: ML ML Alexander Kotelnikov [EMAIL PROTECTED] writes: I have faced a problem with booting from two disks. The trouble seems to be in Herbert's kernel, since mine boots and loads debian-installer initrd image. ML CONFIG_BLK_DEV_FD=m ML ML I guess that is the problem. Ah.. almost sure it is. What boot loader are you using? It should be loading the initrd, not the kernel. Shouldn't it be compiled in anyway? People might want to load additional drivers/plugins/whatever from floppy. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#172828: mklibs: loops on mips
Guido Guenther [EMAIL PROTECTED] writes: The man page doesn't mention an option to actually show the unresolved symbols. You can give -v twice. This should probably be a separate option. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: working on the Alpha port
Herbert Xu [EMAIL PROTECTED] writes: On Thu, Dec 12, 2002 at 05:12:06PM +0300, Alexander Kotelnikov wrote: On Thu, 12 Dec 2002 11:24:25 +0100 ML == Mario Lang [EMAIL PROTECTED] wrote: ML ML Alexander Kotelnikov [EMAIL PROTECTED] writes: I have faced a problem with booting from two disks. The trouble seems to be in Herbert's kernel, since mine boots and loads debian-installer initrd image. ML CONFIG_BLK_DEV_FD=m ML ML I guess that is the problem. Ah.. almost sure it is. What boot loader are you using? It should be loading the initrd, not the kernel. Hmm, it looks like aboot doesn't support loading the initrd from a different floppy than the kernel. (I've only looked at the source, though, not really tried it.) So we could either hack aboot, or get back to load_ramdisk. What do you think? BTW, are you planning on creating any Alpha kernel images udebs? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Default disk partition table type?
Petter Reinholdtsen [EMAIL PROTECTED] writes: OK. Is this a good approximation? /* Need to define on a per arch basis */ #if defined(__i386__) # define DISK_LABEL msdos #elif defined(ia64) # define DISK_LABEL msdos #elif defined(hppa) # define DISK_LABEL msdos #elif defined(__mips__) defined(__MIPSEL__) # define DISK_LABEL msdos #elif defined(__mips__) defined(__MIPSEB__) /* Only supported in libparted 1.6.3? */ # define DISK_LABEL mips /* SGI disklabel */ #else /* not __i386__ */ # error Default DISK_LABEL is not known or not supported on this platform #endif /* not __i386__ */ Alpha needs MSDOS or OSF disklables, depening on the firmware used. So there needs to be a runtime decision. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: keyboard
[EMAIL PROTECTED] writes: PREI'm interested in a price of a Turkish keyboard We don't sell Turkish keyboards. In fact, we don't sell anything. See http://www.debian.org/ for what we actually do. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#154169: mklibs: bug re-introduces (symlinks)
Erik Andersen [EMAIL PROTECTED] writes: On Wed Jul 24, 2002 at 11:28:46AM -0700, Matt Kraai wrote: On Wed, Jul 24, 2002 at 07:42:24PM +0200, Thomas Poindessous wrote: Hi, I tried to rebuild d-i and I looked that symlinks of /lib/ld-2.2.5 to /lib/ld-linux.so.2 (and others) was missing. Please remind me: why do we need to preserve these symlinks? The only reason for the symlinks is to make it easy to upgrade libraries -- not something the installer needs to worry about... Sure, just install the libs with their proper .so names and we can save (1 inode + ~4k)/per symlink from the installer rootfs. The symlinks are converted to hardlinks, so they only take up space in the directory, but usually not enough to push it over 1k, so it saves not a single byte. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Helping a port?
Aneesh Kumar K.V [EMAIL PROTECTED] writes: On Mon, 2002-07-22 at 22:32, Falk Hueffner wrote: Wartan Hachaturow [EMAIL PROTECTED] writes: Recently I've got a Alpha (164LX), and now ready to join the effort. Is there some TODO? The most important thing would be to port the next-generation installer system debian-installer to Alpha. It would probably quite a bit of work, though. Is there any document that explaining how to start hacking. I checked out the code from and I found many design.txt and other .txt files that is really helpful. What i am looking for is a document that explain how to build and test the debian installer on i386 machine. Is the document already there in the CVS ? I did not find anything, either; the README isn't particularly helpful. Perhaps somebody from debian-boot, which I cced, can help? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#152235: no manual page
Tollef Fog Heen [EMAIL PROTECTED] writes: * Philip Blundell | On Mon, 2002-07-08 at 08:52, Tollef Fog Heen wrote: | * Matt Kraai | | | There is no manual page for mklibs. I stumbled across one at | | DebConf2, however, and have appended it below. | | Thanks -- fixed version uploading now. | | Would you care to check the changes you made to mklibs into CVS? | cvs.debian.org:/cvs/debian-boot, module mklibs. sure, I wasn't aware that mklibs was in CVS. Thanks for the heads-up. :) Wouldn't it make more sense to remove it from CVS and make it a build dependency? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: b-f bug maint and reassigning to install-doc
Colin Watson [EMAIL PROTECTED] writes: On 09 Apr 2002 at 10:07:38 -0400, Adam Di Carlo wrote: reassign 126030 milo [...] Bug 126030 must be a milo issue, right? Where does milo come from, anyway? #126030 isn't assigned to anyone at the moment because there is no milo package ... is it hardwired into the kernels or into boot-floppies or something? No, we download the images from SuSE while building... Compiling them is nontrivial, and I don't use MILO, so I didn't feel like making a package. In the long run, that would certainly be better, though... As for the bug report, I can perhaps kludge it to download the potato image. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#147841: Alphaserver 2100 5/250
reassign 147841 kernel-image-2.2.20-generic thanks Looks like a problem with the kernel. Did you ever get any other Linux kernel to boot on this machine? A 2.4 kernel for example? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: bootable woody alpha netinst cd/iso
Ted Goodridge, Jr [EMAIL PROTECTED] writes: I'm offlist, so please CC me directly I am in dire need of a woody bootable iso image for alpha. My floppy drives have died on me and have no way of doing a clean install. The only working os I have is windows Could someone give me a hand? If you've got SRM, and you have network so just the installation system suffices, I could try to quickly kludge an ISO image... But will the potato images not do? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: bootable woody alpha netinst cd/iso
Ted Goodridge, Jr [EMAIL PROTECTED] writes: Um, the potato images will work? How? Well, by burning a CD from them, inserting it into the CDROM drive and booting from it... I'm afraid I cannot really help you unless you're a bit more specific about your problems :) Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: bootable woody alpha netinst cd/iso
Ted Goodridge, Jr [EMAIL PROTECTED] writes: I need to know what to tell the installer (what URL) to grab alpha woody I don't think you can do that... You'll have to install potato first and then upgrade to woody. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Debian GNU/Linux on a floppyless Alpha
Frank Ecke [EMAIL PROTECTED] writes: I've had some fun trying to install the Potato release of Debian GNU/Linux on a DEC Alpha Station. Everything was fine until the installation routine asked me to place a floppy disk into the drive associated with /dev/fd0. Apparently, it was about to create a rescue floppy. I didn't want to do that but had no option to skip that step. What's even worse, even if I had wanted to create such a floppy, I would have been unable to do so. This jewel of an Alpha doesn't have a floppy disk drive, yet the installation routine was convinced of the contrary! The favor I'd like to ask of you now is: Could you modify the installation routine so that those of us who don't want a rescue floppy or don't have a floppy drive can get by? The current version of the boot floppies doesn't even generate a rescue floppy by default, and I was pretty sure the potato version did the same... are you sure it didn't ask for the drivers disk or something? Also it would be helpful to know what firmware you were using and how you booted the machine (tftp?). In any case, the potato boot floppies are unmaintained, and you should probably rather try the pre-release woody boot floppies. If that still doesn't work, please file a bug report. Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: alpha/milo install on multia
Florian Lohoff [EMAIL PROTECTED] writes: except the sparse ext2 which i did not get working (Does the 2.2 Milo really support this) Hmm, I was pretty sure it does. there seems to be a problem with Make system bootable from disk. Selecting this pops up something like Found DOS partition - Acknowledging this immeatly goes to the initial dbootstrap Welcome screen. That probably means dbootstrap segfaulted or similar... please file a bug report. There's a make log floppy option or similar, perhaps you can find something in the logs. No dos filesystem gets created nor does the milo/linload.exe etc get copied to that partition. DOS filesystems currently never get created by the installer, the user always needs to do it by hand :( I was unsuccessful in letting milo boot from the sparse ext2 filesystem - Selecting the partition with hda6: and then issueing an ls within milo showed an empty directory. A boot command like boot hda6:vmlinuz root=/dev/hda6 fails. Weird. Did anybody ever get this to work? As a workaround i copied the kernel to the dos partition which is not really satisfying ... Actually, I've pondered whether it might not be a good idea to have this setup (kernel on FAT next to MILO) as standard. No more problems with MILO incompatibilities... The only drawback I see is that you need to give your kernel images 8+3 names. Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: 3.0.21 bf2.4 report
Othmar Pasteka [EMAIL PROTECTED] writes: mounted perhaps eingebunden for consistency? yeah, that's sounds great, and it makes sense, opposed to the above... i never ever read wurzel-dateisystem in a german book :). I think the standard German term is eingehÃĪngt, which IMHO is also better. It would probably be a good idea to check how SuSE translate this stuff... Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: alpha/milo sparse ext2 install
Florian Lohoff [EMAIL PROTECTED] writes: it seems the Alpha install disks when installing via Milo (i did so on my Multia) does not ask for ext2 kernel 2.0 backward compatibility as it assumes the milo is built on 2.0 kernels (Which was true for potato). Hm, I'm not quite sure I understand what you mean. I changed the logic in 3.0.21, I supposed this is what you tried? The question should always be asked; the default should be yes for MILO and APB and no for SRM. So you didn't see this question? Then there's a bug somewhere With woody it seems the milo is based on 2.2.16-SuSe so it should be feasible to enable sparse ext2 for Alpha/Milo. Unfortunately, the 2.2 series doesn't seem to work for everybody... OK, a cleanup suggestion: - SRM: Never ask, just use new ext2. Nobody uses 2.0 kernels anymore. - MILO/APB: ask if user wants to boot from this partition using ancient MILO/APB. If yes, enable old ext2 and force blocksize to 1k if yes. Default to no. If this sounds reasonable, I will implement it after easter, OK? Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Small patch (makes root.bin about 3900 byte smaller)
Christian Leber [EMAIL PROTECTED] writes:
Hello,
could somebody please apply this small non intrusive patch or say me why
it is not an option?
(against CVS from _now_, it belongs just before the disk loop image
filesystem is created)
Index: rootdisk.sh
===
RCS file: /cvs/debian-boot/boot-floppies/rootdisk.sh,v
retrieving revision 1.344
diff -r1.344 rootdisk.sh
906a907,909
# compress devices.tar.gz better, saves about 3900 byte on the the root image
(cd $R/usr/lib/debootstrap ; mv devices.tar.gz _devices.tar.gz ; zcat
_devices.tar.gz|gzip -9devices.tar.gz ; rm _devices.tar.gz)
I think it would make more sense to create it like that in the first
place (untested):
diff -u -p -r1.344 rootdisk.sh
--- rootdisk.sh 2002/03/06 07:21:05 1.344
+++ rootdisk.sh 2002/03/06 14:00:11
@@ -359,7 +359,7 @@ else
for i in 0 1 2 3 4 5 6 7; do mknod $R/dev/ubd$i b 98 $i; done
fi
- (cd $R ; tar zcf $devtarball dev)
+ (cd $R ; GZIP=--best tar zcf $devtarball dev)
fi
T=${tmpdir}/boot-floppies/prototype-tmp-$$
Falk
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Bug#136561: joel@diggy.com.au: missing drivers in jensen (alpha) boot floppies]
Philip Blundell [EMAIL PROTECTED] writes: Has anybody actually stepped forward to do an alpha build? I don't think we should mark bugs as pending until they really are being worked on. FWIW, I can build some alpha disks today if nobody else is on the case. I'm currently working on this, but I don't have continuous net access, so I don't know how far I'll get. Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Kernel Config's for Alpha Boot Disk's
campbell steven [EMAIL PROTECTED] writes: I've had a pretty good look around to try and find the kernel config file for the debian alpha boot disks, they don't seem to be in the same place like the x86 ones are. If someone can point me in the right direction to find them then that would be great. The kernels are taken from the packages kernel-image-2.2.20-generic, kernel-image-2.2.20-jensen and kernel-image-2.2.20-nautilus, which contain the config. Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#126030: boot-floppies eb164 milo hangs when lanuched by ARC linload.exe
Hi, does any of the 2.2 milos from http://www.suse.de/~stepan/ work? I would like to use them by default, because 2.0 milos can't load kernels from newer ext2 file systems; but if it just doesn't work, I'll try to hardcode 2.0 milo for the release and document this. Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: alpha netboot status?
Philip Blundell [EMAIL PROTECTED] writes: The todo file mentions: o Alpha todo: * fix netboot Anyone know what the specific problem is here? Netboot seems to work OK for me at the moment. I added that ages ago when the netboot stuff didn't even build, I've fixed it since but couldn't test it. So I guess this can be removed now. Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: 3.0.17 alpha nfs-root install problems
Erik Andersen [EMAIL PROTECTED] writes: On Tue Dec 18, 2001 at 09:21:53PM +0100, Falk Hueffner wrote: Just checking out boot-floppies out of CVS and building should work (although it downloads huge amounts of stuff), except currently the busybox package is broken and a new version needs to be built. Care to elaborate on what is wrong with the busybox package? No idea, but there's a version check in the build scripts that complains. So whateve it is, it is fixed already, I'm just waiting for the buildd to catch up. Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: 3.0.17 alpha nfs-root install problems
Sam Johnston [EMAIL PROTECTED] writes: Most impressed with the new installer. Just having some grief with installing to an alpha (specifically multia) which has no disk devices. Boots fine. Mounts root over NFS fine. Downloads files from the network fine. Just refuses to mount the rescue disk In messages I'm seeing things like this: installing /tmp/rescue.bin Mounting /dev/ram on /floppy failed: no such device Mounting /dev/ram on /floppy failed: no such device Mounting /dev/ram on /floppy failed: invalid argument Mounting /dev/ram on /floppy failed: invalid argument Mounting /dev/ram on /floppy failed: no such device Unable to mount the rescue floppy. You may have inserted the wrong floppy. Please try again. Incidentally, cat /target/tmp/tmp/rescue.bin /dev/ram followed by mount -t ext2 /dev/ram /floppy works fine. Hm, that's weird. I'll check it when I get around. Any ideas what's going on here? If it's fixed in a new version, how can I create a new tftpboot.img? Does anyone have such an animal ready to roll? Just checking out boot-floppies out of CVS and building should work (although it downloads huge amounts of stuff), except currently the busybox package is broken and a new version needs to be built. Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: woody install on alpha
Robert Lemmen [EMAIL PROTECTED] writes: i just installed woody on an alpha system and i encountered a couple of potential issues/problems that you might want to hear about: my system is 164ux based and uses arcbios. i downloaded the floppy images and milo from 3.0.17-2001-11-2. - the ruffian milo doesn't seem to be able to read from fat filesystems, and therefore can't boot the rescue floppy. i replaced it with a version i ripped of a suse cd that can read fat. this one says v1.1 Jay's hacked version on startup. Are you sure you're talking about the MILO and not ldmilo.exe? What is the error message you get? - the driver-1.bin and driver-2.bin seem to be corrupt, since they cannot be read by dbootstrap. the md5sums are ok, and the rescue floppy can be read too. Hm, weird. What exactly does dbootstrap say? Perhaps you could try using the report option, which will write a log to a floppy. if i dd these files to floppies, these floppies don't seem to have a filesystem known to me/my kernel any more. i did the installation from net and didn't need the floppies any more. That's normal, they just contain a splitted tar with a few bytes header on each disk. Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#121110: boot-floppies: TFTP boot on Alpha broken
Adam Di Carlo [EMAIL PROTECTED] writes: Falk Hueffner [EMAIL PROTECTED] writes: Package: boot-floppies Version: N/A; reported 2001-11-25 Severity: normal Installing via tftpboot.img will stall at installation of libc6.1, as has been reported to debian-alpha several times, e.g. http://lists.debian.org/debian-alpha/2001/debian-alpha-200110/msg00010.html. rgh told me on IRC: rgh mellum: re init q, when I had that problem ia64, it was because I forgot to boot with root=/dev/ram rgh mellum: without that, the kernel wants to do it's initrd magic, and runs /linuxrc before starting sbin/init. linuxrc is linked to init, so effectively init gets run as pid 12 (or something 1, anyway) rgh linuxrc is run with a pid 1, and the kernel expects that to return before it starts init as pid 1 rgh root=/dev/ram makes the kernel realise that the ramdisk si a real ramdisk, and not an initrd, so it disables the running of linuxrc So we need to find a way to force this parameter. Isn't this the same as #116829 and wasn't it fixed in sysvinit 2.84-1 ? Hm, yes, seems to be the same problem. I'll try to find a tester, since netboot doesn't work for me. Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#121200: Troubles
Alexander Zavaruhin [EMAIL PROTECTED] writes: Package: boot-floppies Version: 2.2.20-2000-12-03 architecture: alpha model: DEC 3000 memory: 32Mb I have boot from SRM console by command b dka500 -fl 0 Floppy was created by command dd if=rescue.bin of=/dev/fd0 rescue.bin - system type jensen Hm, on http://www.debian.org/ports/alpha/sys_types, DEC 3000 are listed as unsupported. Are you sure this is a jensen (DEC 2000)? In any case, there isn't really any work being done with the potato boot floppies any more, so you should rather try the woody boot floppies at ftp://ftp.debian.org/debian/dists/woody/main/disks-alpha/current. We've had at least one report of them working on Jensen (http://lists.debian.org/debian-alpha/2001/debian-alpha-200108/msg00174.html) Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#121110: boot-floppies: TFTP boot on Alpha broken
Package: boot-floppies Version: N/A; reported 2001-11-25 Severity: normal Installing via tftpboot.img will stall at installation of libc6.1, as has been reported to debian-alpha several times, e.g. http://lists.debian.org/debian-alpha/2001/debian-alpha-200110/msg00010.html. rgh told me on IRC: rgh mellum: re init q, when I had that problem ia64, it was because I forgot to boot with root=/dev/ram rgh mellum: without that, the kernel wants to do it's initrd magic, and runs /linuxrc before starting sbin/init. linuxrc is linked to init, so effectively init gets run as pid 12 (or something 1, anyway) rgh linuxrc is run with a pid 1, and the kernel expects that to return before it starts init as pid 1 rgh root=/dev/ram makes the kernel realise that the ramdisk si a real ramdisk, and not an initrd, so it disables the running of linuxrc So we need to find a way to force this parameter. Falk -- System Information Debian Release: testing/unstable Architecture: alpha Kernel: Linux borkum 2.4.15-pre6 #2 Wed Nov 21 19:23:16 CET 2001 alpha Locale: LANG=C, LC_CTYPE=de_DE -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: testing CD that boots
A Mennucc1 [EMAIL PROTECTED] writes: I have a local testing repository and I periodically make CDs out of it, using debian-cd I would like to make a CD that boots, and to use it to install a PC; and to use (and test) the latest boot method what should I do? Install the debian-cd package or check out debian-cd from CVS. Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
langs.xml error
Hi, with CVS boot-floppies I get on Alpha: [...] iconv -f `head -1 croatian.src | sed -e 's/^.*charset=\([^]*\).*/\1/'` -t utf-8 croatian.src croatian.xml iconv -f `head -1 norwegian.src | sed -e 's/^.*charset=\([^]*\).*/\1/'` -t utf-8 norwegian.src norwegian.xml gcc -c -Wall -g -DHAVE_UTF_CONSOLE -I/usr/include/python2.1 iconv.c In file included from /usr/lib/gcc-lib/alpha-linux/2.95.4/include/stdarg.h:36, from /usr/include/python2.1/modsupport.h:10, from /usr/include/python2.1/Python.h:91, from iconv.c:22: /usr/lib/gcc-lib/alpha-linux/2.95.4/include/va-alpha.h:36: warning: redefinition of `va_list' /usr/include/stdio.h:70: warning: `va_list' previously declared here gcc -shared -o iconv.so iconv.o -lpython2.1 ./ver2.py 1 alpha langs.xml processing langs.xml, utf=1, arch=alpha No items for arch alpha found. No file generated make[5]: *** [langs.c] Error 2 make[5]: Leaving directory `/data/home/falk/debian/boot-floppies/boot-floppies/utilities/dbootstrap/langs' What's the problem there? Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: langs.xml error
Claus Hindsgaul [EMAIL PROTECTED] writes: From: Falk Hueffner [EMAIL PROTECTED] Sent: Tuesday, November 20, 2001 4:00 PM What's the problem there? I believe this is because alpha has no items in the *.src files. Look at the english.src. It contains items for all the other architectures in tags such as: item arch=i386 powerpc arm locale=en_GB acm=iso01 font=LatArCyrHeb-16 keymap=i386/qwerty/uk msgcat=en I guess you need to add a line with arch=alpha. See README-Translators for details on the rest of the fields. When a working entry in english.src have been worked out, the rest of the languages' *.src have to construct their variants of the entry. Well, Alpha keyboards aren't really different from i386 keyboards anywhere. Is there a method to express that without exchanging i386 with i386 alpha everywhere? Well, if I change it in english.src, I get: make[5]: Leaving directory `/data/home/falk/debian/boot-floppies/boot-floppies/utilities/dbootstrap/langs' cat po.utf rn.utf ../dbootstrap/langs/english.xml ../dbootstrap/langs/russian.xml ../dbootstrap/langs/polish.xml ../dbootstrap/langs/portuguese.xml ../dbootstrap/langs/french.xml ../dbootstrap/langs/czech.xml ../dbootstrap/langs/german.xml ../dbootstrap/langs/hungarian.xml ../dbootstrap/langs/turkish.xml ../dbootstrap/langs/japanese.xml ../dbootstrap/langs/swedish.xml ../dbootstrap/langs/chinese_t.xml ../dbootstrap/langs/chinese_s.xml ../dbootstrap/langs/danish.xml ../dbootstrap/langs/galician.xml ../dbootstrap/langs/korean.xml ../dbootstrap/langs/croatian.xml ../dbootstrap/langs/norwegian.xml all.utf LC_CTYPE=en_IN.UTF-8 reduce-font /usr/src/unifont.bdf all.utf unifont-reduced.bdf Unable to set locale make[4]: *** [unifont-reduced.bdf] Error 1 make[4]: Leaving directory `/data/home/falk/debian/boot-floppies/boot-floppies/utilities/bogl' Hm, it seems one needs en_IN UTF-8 or something in /etc/locale.gen... Can we avoid that? If not, we should give at least a proper error message. Then I get: LC_CTYPE=en_IN.UTF-8 reduce-font /usr/src/unifont.bdf all.utf unifont-reduced.bdf setlocale: en_IN.UTF-8 FYI: MB_CUR_MAX/MB_LEN_MAX: 6/16 Used chars: 1729 (4112912 processed) bdftobogl -b unifont-reduced.bdf unifont-reduced.bgf bogl_write_font failed make[4]: *** [unifont-reduced.bgf] Error 1 make[4]: Leaving directory `/data/home/falk/debian/boot-floppies/boot-floppies/utilities/bogl' Oh well. I think I'm not in the mood to trace this any further :-/ Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: langs.xml error
Marcin Owsiany [EMAIL PROTECTED] writes: On Tue, Nov 20, 2001 at 06:14:40PM +0100, Falk Hueffner wrote: LC_CTYPE=en_IN.UTF-8 reduce-font /usr/src/unifont.bdf all.utf unifont-reduced.bdf setlocale: en_IN.UTF-8 FYI: MB_CUR_MAX/MB_LEN_MAX: 6/16 Used chars: 1729 (4112912 processed) bdftobogl -b unifont-reduced.bdf unifont-reduced.bgf bogl_write_font failed Well, reduce-font succeeded. But I have never seen that bdftobogl error... Could it be that you ran out of disk space? No. This seems to be a bug in bdftobogl. Adding a bit debug shows: bdftobogl -b unifont-reduced.bdf unifont-reduced.bgf my_write(1, 0x120008846, 4) write(1, 0x120008846, 4) - 4 write(1, 0x12000884a, 0) - 4 my_write(1, 0x111e0, 48) write(1, 0x111e0, 48) - 48 write(1, 0x11210, 0) - 48 my_write(1, 0x12001a990, 16) write(1, 0x12001a990, 16) - 16 write(1, 0x12001a9a0, 0) - 16 my_write(1, 0x1200664c0, 1024) write(1, 0x1200664c0, 1024) - 1024 write(1, 0x1200668c0, 0) - 1024 index_size = 3714 my_write(1, 0x1200668d0, 29712) write(1, 0x1200668d0, 29712) - 15292 my_write: Bad address write(1, 0x12006a48c, 14420) - -1 bogl_write_font failed 5 Probably in bdftobogl.c, if (my_write(fd, font-index, index_size * sizeof(int *))) should really be if (my_write(fd, font-index, index_size * sizeof(int))) (the bogl author seems to be of the long = make sure it's really 32 bit fraction. Sigh.) Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: test of woody tftpboot.img stalls
(I cc this to debian-boot in case somebody has an idea about the init stuff) Mikael Westerberg [EMAIL PROTECTED] writes: Still problems with latest tftboot.img(3.0.16-2001-10-30). During two attempts to install the installation has stalled while extracting libc6. I can still switch consoles and the maschine still responds to pings, but I can not activate the shell on the second console. Hm, there seems to be some problem in the libc postinst; we've had another similar report before from an AlphaStation 500/500 user. I've never seen this, though. Could you perhaps paste the last few lines of output of the postinst script? It should appear on console 4. The postinst contains this: # Black magic. If we are in a chroot, then /proc/1/exe will not resolve to # a file (will return EPERM). This keeps us from breaking things in # debootstrap, and the like. case $sysvinitver in $badsysvinitver) ;; *) if [ -x /sbin/init -a -x /bin/readlink ]; then if [ $(readlink /proc/1/exe 2/dev/null) = /sbin/init ]; then (init u ; sleep 1) fi fi ;; esac Perhaps this magic fails, and it thinks it needs to restart init? This would explain what you see. Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Cant install anymore... VERY CRITICAL
Roger Eklund [EMAIL PROTECTED] writes: its impossible to install debian from the net... ive tried woody several times and several days now on diffrent connections and computers... its something in the installation scrip.. it downloads the base system and extracts it.. but when it should install it.. the comman is chroot dpkg --force... I get an error.. plzz fix.. =) There are move verbose message on another console, I think it's the 4th. So press Alt-F4 and check the messages there. Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Woody install report
Ethan Benson [EMAIL PROTECTED] writes: (3) in the post-boot configuration, why are MD5 passwords defaulted to No ie not enabled ? compatibility with crufty old unixes i suppose. if you were to make the box into an NIS server that would matter i think. Somebody who wants that will know to what to answer. People who don't know what this is all about will want MD5 passwords, so this should be the default. Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#116413: Alpha: Wrong partition No. for aboot
Jan-Benedict Glaw [EMAIL PROTECTED] writes: Package: boot-floppies Architecture: Alpha Version: current/ Distribution: woody/sid After installing my videocard-less system (with more or less PITA) I had difficulties to boot it. I've added an extra partition for /boot (as I normally do for all other architectures I am workin on as well), but this wasn't recognized by dbootstrap. dbootstrap set my root partition's partition number as the partition ID where to search /boot/vmlinuz-2.2.19-generic in. Wrong. Ok, I can fix this, the SILO part already has equivalent code to handle this that I can steal. But I don't see any point in having a separate /boot partition. Why do you think a special partition would be useful? Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: installing
ivo jorris graad.p [EMAIL PROTECTED] writes: I bought a set of Debian 2.2r3. there are 3 source cds, 3 binary alpha cds and 3 binary i386 cds. I want to install Debian into DEC Personal Workstation 600a, I've upgrade the bios and now I'm using SRM console. I want to ask how to install Debian into my workstastion? I've read your site about installing Debian for alpha, but I still can't understand how to make bootdisk and what is the purpose of source cd, binary alpha cd, binary i386 cd?? Please tell what should I write to boot from CD-ROM by SRM console. You need to find out the SRM device name with show dev. Then type 'boot xxx -fl 0' where xxx is your CD-ROM device. If this doesn't work, you might also want to try the upcoming woody boot disks from ftp.debian.org/debian/dists/woody/main/disks-alpha/current, where a README.txt also explains how to create boot floppies. Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Alpha Milo/rescue/root floppies on LX164
[EMAIL PROTECTED] writes: With ArcBIOS v5.70: The LX164 MILO disk works great. The install is almost flawless, with two exceptions: tasksel exits with an error stating that all task definitions are duplicated. This condition is present also in the woody sid versions of tasksel on a running system. Hmm, could you please file a bug report against tasksel about this? The installer will not write linload.exe and milo to the 5M FAT16 partition (/dev/hda1). This may be due to my own incompetence, as it's never worked for me. Hm, was there any reason in the log? You can use the report bug or similar menu entry to save the logs to the floppy. I've not yet tried MILO installation, but it *should* work :) However, going to VT2, I find that there is no provision in the install to make a FAT16 filesystem. A clue? There's a mkdosfs on the boot floppy, but no menu entry to access it so you'd need to use the shell IIRC. I will document this for now, and see if I can hack it to be accessible from the menu. BTW, does anybody know if it is possible to launch AlphaBios from SRM on an SX164? I know how to *install* it, but for testing purposes it would be nice if I could launch it off a floppy. With SRM v5.8-1: Boots fine from the rescue floppy, installs fine (with the tasksel exception) and then the installed system won't boot. The kernel loads and starts, then gets an error -- ID 1 is respawning too fast, disabling for 5 minutes. And that's that. Weird, I have no idea about this... ID 1 would be init itself?? I'm not sure that the size of the disk isn't the problem. I've sort of assumed that the x86 1023-cylinder-syndrome doesn't apply in AXP-land. I don't think aboot has any limit like that... Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Alpha Boot Floppies: new version
Doug Larrick [EMAIL PROTECTED] writes:
The good news? bootp/tftp boot works great! In SRM, you need to
select bootp via console variable ('set ewa0_protocol bootp' or
something like that) rather than via a flag to the boot command
('boot -proto bootp ewa0' does not work, failing with Block
FFF2 is not in any zone messages).
That's really good news. I couldn't get bootp to work, so I couldn't
test it. There were some suspicious hacks in the previous version...
The bad news? The installation hangs during/right after installing
the libc6.1 package. It hangs so solid that the shell on the second
virtual console will not respond. I even tried floppies rather than
tftp and it stops in the same location. Is this a known problem?
No. That's really weird; I have no idea what might cause this. I had
no problems at that point. Did you download via http, or did you
install from a mounted mirror?
Also, when trying to write aboot to the disk, it looks like the
installer is trying to swriteboot to the partition where the root
filesystem is (/dev/sdan) rather than directly to /dev/sda .
I noticed this too. If this is not just the message being wrong, this
might even destroy data... I'll check it.
Falk
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Alpha Boot Floppies: new version
Doug Larrick [EMAIL PROTECTED] writes: Quoting Falk Hueffner [EMAIL PROTECTED]: That's really good news. I couldn't get bootp to work, so I couldn't test it. There were some suspicious hacks in the previous version... Bootp is a lot easier to get going using a dhcp server rather than a bootp server. Much better log messages. Or did you mean the SRM end? Yes, I meant the SRM end. I couldn't get SRM tro transmit *anything* over ethernet (checked with tcpdump), regardless of the medium type setting. Perhaps this is because I had a crappy ISA card at the other end; they can communicate from Linux, though. The bad news? The installation hangs during/right after installing the libc6.1 package. It hangs so solid that the shell on the second virtual console will not respond. I even tried floppies rather than tftp and it stops in the same location. Is this a known problem? No. That's really weird; I have no idea what might cause this. I had no problems at that point. Did you download via http, or did you install from a mounted mirror? Downloaded via http. If there's an easy way to make a local mirror of just the files needed for the base system I'd be willing to give that a try. Anthony Towns just posted a method on debian-boot: /usr/sbin/debootstrap --download-only --arch alpha woody woody-alpha http://ftp.debian.org/debian/ (cd woody-alpha tar cf ../basedebs_3.0_alpha.tar *) rm -rf woody-alpha Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Alpha Boot Floppies: new version
[EMAIL PROTECTED] writes: Anthony Towns just posted a method on debian-boot [of making a basedebs.tgz file] OK, I gave this a try, same result; let me type in the end of console 4 (anything really odd is probably a typo): dpkg: warning - ignoring pre-depencency problem ! (Reading database... 77 files and directories currently installed.) Preparing to replace dpkg 1.9.16 (using ../archives/dpkg_1.9.16_alpha.deb) ... Unpacking replacement dpkg ... dpkg: dpkg: dependency problems, but configuring anyway as you request: dpkg depends on libc6.1 (= 2.2.3-1); however: Package libc6.1 is not installed. dpkg depends on libncurses5 (= 5.2.20010310-1); however: Package libncurses5 is not installed. dpkg depends on libstdc++2.10-glibc2.2; however: Package libstdc++2.10-glibc2.2 is not installed. Setting up dpkg (1.9.16) ... Selecting previously deselected package libc6.1. (Reading database ... 216 files and directories currently installed.) Unpacking libc6.1 (from ../libc6.1_2.2.3-8_alpha.deb) ... Setting up libc6.1 (2.2.3-7) ... Current default timezone: 'UTC'. Local time is now: Tue Aug 14 20:07:43 UTC 2001. Universal time is now: Tue Aug 14 20:07:43 UTC 2001. Run 'tzconfig' if you wish to change it. Package 'sysvinit' is not installed and no info is available. Use dpkg --info (= dpkg-deb --info) to examine archive files, and dpkg --conents (= dpkg-deb --contents) to list their contents. ... and that's all she wrote, hard hang at this point (for those who arrived late, this is on a Bret, aka AlphaStation 500/500). Perhaps it's trying to restart init then in a bad way. I'll retrieve the postinst and have a look later. Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Alpha bootdisk status
Istvan Gyenes [EMAIL PROTECTED] writes: On 8 Aug 2001, Falk Hueffner wrote: Here are my plans for the Alpha bootdisks: * special Jensen and Nautilus stuff. I can't test this, I can only try not to break it. I'll try to test the Jensen specific booting. As soon as I have time. Maybe next week, hope it is not too late. That would be great! Q: Is the EV4 patches applyed to the kernel for the Jensen (traps.c patch)? I don't know; the bootfloppies simply use the kernel from the kernel-image-2.2.19-jensen package. If you find it is missing an important patch, please make a bugreport against it. Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Alpha Boot Floppies: new version
Hi, I've built new boot floppies from the current boot-floppies CVS and put them on ftp://134.2.15.134/boot-floppies/disks-alpha/. I think they are now in a state to be added to ftp.debian.org at the next boot-floppies release. I've used MILO 2.2-17 this time, since 2.2-18 was reported to be broken on at least two subarchitectures. Unfortunately, I have only my SX164 to test, so I'd appreciate any reports. I have not yet tested the complete installation on my machine, since I have pppoe for internet access which isn't supported. make release now works fine except for two things: * You need the srmbootfat tool from the MILO source. In the long run, this could be provided by a MILO deb package. Currently, we could either require it to be locally installed, or I could add it to the CVS. It's two files of 12k (GPL). If this is considered a good idea, I'd like to create an alpha-specials dir. Any opinions? * Downloading via apt-get still doesn't work because of dependency problems. Could I perhaps just commit the fake-get hack which uses wget instead and activate it for Alpha? Even if we fiddle with the packages, this problem can always reappear. Hmm, that's it for now :) Next thing will be to fix netboot. We could also need a tester for nautilus, since I have no idea about the current state. Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Alpha bootdisk status
Hi, Herbert Xu has uploaded a fixed ash to incoming, so the crash problem when installing modules will go away soon. The alpha bootdisk generation is generally a bit crufty and broken, I intended to clean up things a bit, but I have not much knowledge about the different Alpha systems, I've only ever tried SRM and AlphaBios. Here are my plans for the Alpha bootdisks: * kernel floppy (rescue is a bit misleading here). Ext2. Contains a generic 2.2.19 kernel. Bootable directly by SRM consoles via aboot. Bootable from other consoles via milo disks. Contains extra UP 1000 aka Nautilus stuff (how do UP 1000's actually boot?). Tested for SRM already. Mostly completed. * 19 milo_$SUBARCH disks. Each contains a subarch-specific milo, and a loadlin.exe/ldmilo.exe, and has been prepared to be booted from SRM with srmbootfat (though using the aboot on the kernel floppy is preferable). The milo binaries will be downloaded from http://www.suse.de/~stepan/binaries/2.2-18/ with wget (I'd add a builddep). loadlin.exe/ldmilo.exe will be downloaded from ftp.debian.org, I don't know if there is any more 'generic' place. Problem here: srmbootfat is not available anywhere as binary. This is the only step that can't be made automatically easily: it needs to be extracted from http://www.suse.de/~stepan/source/milo-2.2-18.tar.bz2 and built manually. David Huggins-Daines has already written a script to build milo disks, I'll only have to activate that. Question: did I get it right that Ruffian needs ldmilo.exe and all other loadlin.exe? I couldn't find documentation on that... Note in the past we provided only milo binaries, but I find milo disks more convenient, and they can also be booted from SRM which is pointless for the install, but might be handy for rescue since MILO knows devices SRM can't access. * root disk. Contains gzipped ext2 fs directly written to floppy. Barely fits now. Completed except for bug fixes. * special Jensen and Nautilus stuff. I can't test this, I can only try not to break it. * remove support for disks with special kernel images for the subarch's (except jensen and nautilus, which seem to be unable to run generic kernels). Would be nice if a kernel would fit together with a milo on a disk, but it doesn't. We also don't have those kernel images in the archive anymore. * net boot stuff. I'll disable it until I get around to understanding it. Well, I hope I can get a bit of feedback about this plan, especially on the milo stuff and the intent to remove subarch kernel support... Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Library reduction on Alpha (was: Re: Alpha: aboot rescue disk)
Adam Di Carlo [EMAIL PROTECTED] writes: P. S. the mklibs.sh libm trick didn't seem to work Meaning -- didn't reduce far enough? Yes, it acually doesn't seem to make any difference. With mklibs.sh: -rw-r--r--1 root root 1743331 Aug 7 18:37 root.bin total 4273 -rwxr-xr-x2 root root 124160 Aug 7 18:36 ld-2.2.3.so* -rwxr-xr-x2 root root 124160 Aug 7 18:36 ld-linux.so.2* -rw-r--r--2 root root 963944 Aug 7 18:36 libc-2.2.3.so -rw-r--r--2 root root 963944 Aug 7 18:36 libc.so.6.1 -rw-r--r--2 root root Aug 7 18:34 libcom_err.so.2 -rw-r--r--2 root root Aug 7 18:34 libcom_err.so.2.0 -rw-r--r--2 root root14608 Aug 7 18:34 libdl-2.2.3.so -rw-r--r--2 root root14608 Aug 7 18:34 libdl.so.2.1 -rw-r--r--2 root root22904 Aug 7 18:34 libe2p.so.2 -rw-r--r--2 root root22904 Aug 7 18:34 libe2p.so.2.3 -rw-r--r--2 root root90488 Aug 7 18:34 libext2fs.so.2 -rw-r--r--2 root root90488 Aug 7 18:34 libext2fs.so.2.4 -rw-r--r--2 root root 578896 Aug 7 18:34 libm-2.2.3.so -rw-r--r--2 root root 578896 Aug 7 18:34 libm.so.6.1 -rw-r--r--2 root root70800 Aug 7 18:30 libnewt.so.0.50 -rw-r--r--2 root root70800 Aug 7 18:30 libnewt.so.0.50.17 -rw-r--r--2 root root17504 Aug 7 18:26 libnss_dns-2.2.3.so -rw-r--r--2 root root17504 Aug 7 18:26 libnss_dns.so.2 -rw-r--r--2 root root34264 Aug 7 18:29 libpopt.so.0 -rw-r--r--2 root root34264 Aug 7 18:29 libpopt.so.0.0.0 -rw-r--r--2 root root86352 Aug 7 18:29 libresolv-2.2.3.so -rw-r--r--2 root root86352 Aug 7 18:29 libresolv.so.2.1 -rw-r--r--2 root root 132168 Aug 7 18:32 libslang.so.1 -rw-r--r--2 root root 132168 Aug 7 18:32 libslang.so.1.4.4 -rw-r--r--2 root root13488 Aug 7 18:27 libuuid.so.1 -rw-r--r--2 root root13488 Aug 7 18:27 libuuid.so.1.2 With mklibs.py: -rw-r--r--1 root root 1431323 Aug 7 18:56 root.bin total 3083 -rwxr-xr-x2 root root 124160 Aug 7 18:55 ld-2.2.3.so* -rwxr-xr-x2 root root 124160 Aug 7 18:55 ld-linux.so.2* -rw-r--r--2 root root 948288 Aug 7 18:55 libc-2.2.3.so -rw-r--r--2 root root 948288 Aug 7 18:55 libc.so.6.1 -rw-r--r--2 root root Aug 7 18:55 libcom_err.so.2 -rw-r--r--2 root root Aug 7 18:55 libcom_err.so.2.0 -rw-r--r--2 root root14608 Aug 7 18:55 libdl-2.2.3.so -rw-r--r--2 root root14608 Aug 7 18:55 libdl.so.2.1 -rw-r--r--2 root root22904 Aug 7 18:55 libe2p.so.2 -rw-r--r--2 root root22904 Aug 7 18:55 libe2p.so.2.3 -rw-r--r--2 root root90488 Aug 7 18:55 libext2fs.so.2 -rw-r--r--2 root root90488 Aug 7 18:55 libext2fs.so.2.4 -rw-r--r--2 root root 3168 Aug 7 18:55 libm-2.2.3.so -rw-r--r--2 root root 3168 Aug 7 18:55 libm.so.6.1 -rw-r--r--2 root root70784 Aug 7 18:55 libnewt.so.0.50 -rw-r--r--2 root root70784 Aug 7 18:55 libnewt.so.0.50.17 -rw-r--r--2 root root17504 Aug 7 18:52 libnss_dns-2.2.3.so -rw-r--r--2 root root17504 Aug 7 18:52 libnss_dns.so.2 -rw-r--r--2 root root34264 Aug 7 18:55 libpopt.so.0 -rw-r--r--2 root root34264 Aug 7 18:55 libpopt.so.0.0.0 -rw-r--r--2 root root72256 Aug 7 18:55 libresolv-2.2.3.so -rw-r--r--2 root root72256 Aug 7 18:55 libresolv.so.2.1 -rw-r--r--2 root root 132120 Aug 7 18:55 libslang.so.1 -rw-r--r--2 root root 132120 Aug 7 18:55 libslang.so.1.4.4 -rw-r--r--2 root root13488 Aug 7 18:55 libuuid.so.1 -rw-r--r--2 root root13488 Aug 7 18:55 libuuid.so.1.2 so libm is still the culprit. I think I'll just change rootdisk.sh to use mklibs.py for Alpha for now. Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Library reduction on Alpha (was: Re: Alpha: aboot rescue disk)
Christopher C. Chimelis [EMAIL PROTECTED] writes: On 7 Aug 2001, Falk Hueffner wrote: Yes, it acually doesn't seem to make any difference. With mklibs.sh: What exactly does this script do to libc (or really, what isn't it doing now)? There are some binutils issues that I'm working on that may be related... The only difference between the two scripts is how the set of symbols to be incuded is determined. mklibs.py simply adds missing symbols iteratively until everything is satisfied. mklibs.sh does something different (I just realized that it doesn't do what I thought it did... unfortunately there's no real documentation... Hm, I'm afraid somebody else will have to tell you how it works :) You can find both scripts in the boot-floppies CVS, in scripts/rootdisk. Well another difference: mklibs.py uses readelf insead of objdump, since that seemed to be slightly easier to parse. Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: New bootdisks / first install report / help needed
Adam Di Carlo [EMAIL PROTECTED] writes: Helge Kreutzmann [EMAIL PROTECTED] writes: I can try them on a ruffian which is currently out of production due to some technical difficulties and on my LX, maybe on an XP1000. Regarding my LX I cannot complete the installation due to missing network (i.e. only modem). How is a milo based system supposed to be booted ? I see special disk for nautilus and jensen but no for milo based systems. I'd have to leave that to an Alpha porter to answer. I'm afraid the documentation probably won't be much help. I know about two MILO systems: Those with AlphaBios, and the Ruffian, which has an ARCS-Bios (and I actually know only about AlphaBios first-hand). Both understand the FAT filesystem, so you need a FAT-formatted Floppy with the correct milo and linload.exe, which is a tiny program that gets started from the bios and loads the MILO. We could either provide the MILO images and instruct the users to prepare such floppies, or provide complete floppy images. The latter would be more wasteful for the mirrors since the milo only takes about 400k, so we waste about 15MB mirror space, but I'd actually prefer it, makes things easier for the installer, no problems with finding a mkfatfs, getting file names right etc. I wanted to hack that part anyway, OK if I give it a try? Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Alpha: aboot rescue disk
Hi, currently, the rescue disk for Alphas with SRM/aboot doesn't work because e2writeboot fails. The first problem is that e2writeboot doesn't return a nonzero status on failure so this will go unnotified; I've tried filing a bug about this but the BTS is down. It fails because the disk is too full; it needs a continuos empty area and can't find one. First, it might help to install it onto the empty disk, when finding a continuos area is not a problem. I'll look into that. Then I wonder if the sys_map.gz (77k) is actually needed, or was it put there just because it fits so nicely? There's already a special case for some arch's not to add it, can I just include Alpha? Falk P. S. the mklibs.sh libm trick didn't seem to work so everybody building boot disks for Alpha should try mklibs.py, which is in CVS now. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: libm on root-filesystem?
Goswin Brederlow [EMAIL PROTECTED] writes: ls -l lib total 2137 -rwxr-xr-x1 root root 124160 Aug 6 06:22 ld-2.2.3.so* mklibs.py: ls -l lib total 1542 -rw-r--r--1 root root 124160 Aug 6 06:03 ld-2.2.3.so Hmm, needs to be made executable IIRC, but I'm too tired now and will go to bed. Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Build problem for Alpha bootdisks
Adam Di Carlo [EMAIL PROTECTED] writes:
But didn't someone mention this was a sysvinit bug which was getting
fixed?
I have no idea. In any case, as Ethan Benson already pointed out
(http://lists.debian.org/debian-boot-0107/msg00639.html), the current
method is wasteful because it downloads all dependencies. So I suggest
this as a workaround till we can get apt-get to ignore dependencies
*completely*, slightly modified from Ethan's suggestion:
##
## Download a package like apt-get would do; only ignore dependencies.
## This is a hack. Should apt-get ever get an option to do that, it
## should be used instead.
##
fake_get () {
# HACK: we assume the first mirror is correct...
local MIRROR=$(awk '/^deb / {print $2; exit}' sources.list)
for i in $@ ; do
# HACK: we assume the first package is correct...
local PACKAGE=$(apt-cache show $i | awk '/^Filename:/ {print $2; exit}')
debug downloading $MIRROR/$PACKAGE
(cd $archive/cache/archives wget -q $MIRROR/$PACKAGE)
done
}
It would be nice if somebody could try this for another architecture...
Falk
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Build problem for Alpha bootdisks
Thimo Neubauer [EMAIL PROTECTED] writes: On Mon, Jul 30, 2001 at 03:45:37AM +0200, Falk Hueffner wrote: Do I have to add sysvinit to the package list, or what's the problem? The problem was already being discussed starting at http://lists.debian.org/debian-boot-0107/msg00603.html. The solution is too simle: you just have to download some packages manually, i.e. e2fsprogs, which isn't needed for the bootdisk (and won't be unpacked). You'll need to do this until apt-get gets a switch for telling it: Hey, just download the packages without checking deps! I want it! I know what I'm doing! :-) Hm, this sucks. Maybe I'll try to implement the wget idea mentioned in that thread. I also wonder if we shouldn't use the milos from http://www.suse.de/~stepan/, which are actively being maintained by the SuSe folks. If nobody objects, I could try it. That sounds great, because all MILOs I tried so far suck (can't read my ext2 with sparse superblocks, even with a MILO said to be made out of a 2.4.x-kernel...) Huh? I've never heard of any 2.4.x MILO... Well, I'll just try to use stepan's MILO, shouldn't be harder than changing a few paths. Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Build problem for Alpha bootdisks
Hi, I'm just trying to build the boot disks for Alpha, but I get: I: downloading required packages from files /data/src/boot-floppies/scripts/rootdisk/EXTRACT_LIST_all /data/src/boot-floppies/scripts/rootdisk/EXTRACT_LIST_alpha D: ash base-passwd busybox console-data dhcp-client debootstrap e2fsprogs-bf libnewt0 libpopt0 makedev modutils nano-tiny netbase net-tools slang1 whiptail libc6.1 util-linux aboot dosfstools D: download command is apt-get -q --yes -o Dir::Etc::SourceList=/data/src/boot-floppies/sources.list -o Debug::NoLocking=true -o Dir::Cache=/data/mirror/bootfloppies-download/cache -o Dir::State::status=/data/mirror/bootfloppies-download/status -o Dir::State::Lists=/data/mirror/bootfloppies-download/lists -o Dir::Etc::preferences=/data/src/boot-floppies/preferences.apt Hit http://http.us.debian.org unstable/main Packages Hit http://http.us.debian.org unstable/main Release Hit http://http.us.debian.org testing/main Packages Hit http://http.us.debian.org testing/main Release Reading Package Lists... Building Dependency Tree... Reading Package Lists... Building Dependency Tree... D: not downloading, have file '/data/mirror/bootfloppies-download/cache/archives/console-data_1999.08.29-21.1_all.deb' D: not downloading, have file '/data/mirror/bootfloppies-download/cache/archives/libc6.1_2.2.3-7_alpha.deb' D: downloading package(s) ash base-passwd busybox dhcp-client debootstrap e2fsprogs-bf libnewt0 libpopt0 makedev modutils nano-tiny netbase net-tools slang1 whiptail util-linux aboot dosfstools Reading Package Lists... Building Dependency Tree... Some packages could not be installed. This may mean that you have requested an impossible situation or if you are using the unstable distribution that some required packages have not yet been created or been moved out of Incoming. The following information may help to resolve the situation: Sorry, but the following packages have unmet dependencies: modutils: Depends: sysvinit (= 2.71-2) but it is not going to be installed E: Sorry, broken packages can't find package, or no such package, 'ash' can't find package, or no such package, 'base-passwd' can't find package, or no such package, 'busybox' can't find package, or no such package, 'dhcp-client' can't find package, or no such package, 'debootstrap' can't find package, or no such package, 'e2fsprogs-bf' can't find package, or no such package, 'libnewt0' can't find package, or no such package, 'libpopt0' can't find package, or no such package, 'makedev' can't find package, or no such package, 'modutils' can't find package, or no such package, 'nano-tiny' can't find package, or no such package, 'netbase' can't find package, or no such package, 'net-tools' can't find package, or no such package, 'slang1' can't find package, or no such package, 'whiptail' can't find package, or no such package, 'util-linux' can't find package, or no such package, 'aboot' can't find package, or no such package, 'dosfstools' E: ./rootdisk.sh abort make[1]: *** [root.bin] Error 255 make[1]: Leaving directory `/data/src/boot-floppies' make: *** [build] Error 2 Do I have to add sysvinit to the package list, or what's the problem? I also wonder if we shouldn't use the milos from http://www.suse.de/~stepan/, which are actively being maintained by the SuSe folks. If nobody objects, I could try it. Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
