Bug#442103: initramfs-tools: update-initramfs fails with lilo and a degraded raid1 array
clone 442103 -1 -2 reassign -1 lilo retitle -1 lilo.conf provide an option equiv to -H switch severity -1 wishlist reassign -2 lilo retitle -2 lilo usage needs to show -H severity -2 minor stop On Mon, 07 Jan 2008, Ionut Ciocirlan wrote: I think Jens is right. Maybe this should rather be a wishlist against lilo. Regards, Ionut please such easy bug triage can be done by anyone who cares, see bugs.debian.org thanks -- maks -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Processed: Re: Bug#442103: initramfs-tools: update-initramfs fails with lilo and a degraded raid1 array
Processing commands for [EMAIL PROTECTED]: clone 442103 -1 -2 Bug#442103: initramfs-tools: update-initramfs fails with lilo and a degraded raid1 array Bug 442103 cloned as bugs 459713-459714. reassign -1 lilo Bug#459713: initramfs-tools: update-initramfs fails with lilo and a degraded raid1 array Bug reassigned from package `initramfs-tools' to `lilo'. retitle -1 lilo.conf provide an option equiv to -H switch Bug#459713: initramfs-tools: update-initramfs fails with lilo and a degraded raid1 array Changed Bug title to `lilo.conf provide an option equiv to -H switch' from `initramfs-tools: update-initramfs fails with lilo and a degraded raid1 array'. severity -1 wishlist Bug#459713: lilo.conf provide an option equiv to -H switch Severity set to `wishlist' from `wishlist' reassign -2 lilo Bug#459714: initramfs-tools: update-initramfs fails with lilo and a degraded raid1 array Bug reassigned from package `initramfs-tools' to `lilo'. retitle -2 lilo usage needs to show -H Bug#459714: initramfs-tools: update-initramfs fails with lilo and a degraded raid1 array Changed Bug title to `lilo usage needs to show -H' from `initramfs-tools: update-initramfs fails with lilo and a degraded raid1 array'. severity -2 minor Bug#459714: lilo usage needs to show -H Severity set to `minor' from `wishlist' stop Stopping processing here. Please contact me if you need assistance. Debian bug tracking system administrator (administrator, Debian Bugs database) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#442103: initramfs-tools: update-initramfs fails with lilo and a degraded raid1 array
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Fri, Sep 28, 2007 at 11:43:33AM +0200, Jens Thiele wrote: On 17 Sep 2007, [EMAIL PROTECTED] wrote: hey guys sorry if you still use lilo, you need to do manual intervence, i see no other way. unless a real strong argument comes up, i'll close that bug in a month. AFAIK grub does not support LVM on top of md raid1 whereas lilo does? The installer allows to setup LVM on top of raid and as far as I remember then suggests usage of lilo. Just as explanation / usage scenario: I use lvm on top of raid1 on a laptop and run the raid in degraded state most of the time. From time to time I attach an external usb disk and resync it. (incremental resync that is) GRUB does not support LVM or RAID, but your syncing goal can be reach if you store /boot on RAID1 but outside LVM. This is what I use for servers, which should work fine for your use too: /dev/sda - GRUB stage1 /dev/sdb - GRUB stage1 (backup) /dev/sda1 + /dev/sdb1 - /dev/md0 (RAID1) - /boot /dev/sda2 + /dev/sdb2 - /dev/md1 (RAID1) - swap /dev/sda3 = /dev/sdc3 - /dev/md2 (RAID1) - LVM2 - ... The trick used is that data on RAID1 is readable directly without knowledge of RAID, so GRUB reads /boot on /dev/sda1. Just make sure to not use the savedefault feature of GRUB (you don't ever want to _write_ directly to a RAID1 partition!). The grub-install wrapper (or is it install-grub?) does not support this trick (it cannot reliably resolve the underlying partition of a RAID1), but you can do it manually like this: 1) Copy GRUB stage2 and stage3 into mounted /boot RAID1 partition: mkdir /boot/grub cp /usr/lib/grub/i386-pc/* /boot/grub/ 2) Install GRUB stage1 onto bootsector grub ...and then within the GRUB shell: root (hd0,0) setup (hd0) quit Hope that helps (to get rid of that old ugly LILO). - Jonas -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFHgkPyn7DbMsAkQLgRAnZhAJ4rlwiFW4m96Yv/OASCCHGSgtw2SwCbB0XY R7Wg8k0lGju30lhfc6xYsf0= =idza -END PGP SIGNATURE- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#442103: initramfs-tools: update-initramfs fails with lilo and a degraded raid1 array
On 1/6/08, Jens Thiele wrote: Hmm. I think the only sensible solution would be to provide an option in the lilo.conf I think Jens is right. Maybe this should rather be a wishlist against lilo. Regards, Ionut -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#442103: initramfs-tools: update-initramfs fails with lilo and a degraded raid1 array
On Jan 6, 2008 8:15 PM, maximilian attems [EMAIL PROTECTED] wrote: so let lilo die slowly and painful. Until grub manages to boot this here machine... please don't let it die! :-P -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#442103: initramfs-tools: update-initramfs fails with lilo and a degraded raid1 array
maximilian attems [EMAIL PROTECTED] writes: as debian supports partial upgrades and initramfs-tools is not gonna start depend on lilo this inocation can only be used after Lenny release. marking as wontfix. Let me see if I got you right: You want to mark it as wontfix because older versions of lilo do not provide -H = if one adds -H to the lilo invocations in update-initramfs there might be problems during update from etch? Maybe we could test the lilo version and add -H if version some version? But in the end this is not really a good solution because maybe not everyone wants to use -H. Hmm. I think the only sensible solution would be to provide an option in the lilo.conf Sorry for the noise. Greetings Jens -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#442103: initramfs-tools: update-initramfs fails with lilo and a degraded raid1 array
On Sun, Jan 06, 2008 at 04:00:39PM +0100, Jens Thiele wrote: Let me see if I got you right: You want to mark it as wontfix because older versions of lilo do not provide -H = if one adds -H to the lilo invocations in update-initramfs there might be problems during update from etch? Maybe we could test the lilo version and add -H if version some version? But in the end this is not really a good solution because maybe not everyone wants to use -H. Hmm. I think the only sensible solution would be to provide an option in the lilo.conf no noise. yes you got me right, the stable lilo has no -H switch thus you would have to add a versioned depend to make sure that the lilo you are running is new enough. i doubt that any grub/grub2 user might be happy about that. so let lilo die slowly and painful. -- maks -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#442103: initramfs-tools: update-initramfs fails with lilo and a degraded raid1 array
as some of you guys have a real usage scenario for lilo i checked and got zero output: ./lilo --help 21 | grep -- -H lilo does not print that flag yet in it's usage, so get that fixed first then i can use it in update-initramfs. happy hacking -- maks -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#442103: initramfs-tools: update-initramfs fails with lilo and a degraded raid1 array
On Fri, 28 Sep 2007, Jens Thiele wrote: On 17 Sep 2007, [EMAIL PROTECTED] wrote: hey guys sorry if you still use lilo, you need to do manual intervence, i see no other way. unless a real strong argument comes up, i'll close that bug in a month. AFAIK grub does not support LVM on top of md raid1 whereas lilo does? The installer allows to setup LVM on top of raid and as far as I remember then suggests usage of lilo. good reason enough for separate small / or /boot on raid Just as explanation / usage scenario: I use lvm on top of raid1 on a laptop and run the raid in degraded state most of the time. From time to time I attach an external usb disk and resync it. (incremental resync that is) IMHO lilo should provide an option in lilo.conf corresponding -H. See also bug: #278373 i fail to find the mentioned lilo -H mention in the lilo manpage. dpkg -l lilo | grep ^ii ii lilo 22.6.1-9.3 LInux LOader - The Classic OS loader can loa a lilo -H invocation leads into usage() message. what are you guys using? happy hacking -- maks Greetings Jens -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- maks -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#442103: initramfs-tools: update-initramfs fails with lilo and a degraded raid1 array
maximilian attems [EMAIL PROTECTED] writes: i fail to find the mentioned lilo -H mention in the lilo manpage. dpkg -l lilo | grep ^ii ii lilo 22.6.1-9.3 LInux LOader - The Classic OS loader can loa a lilo -H invocation leads into usage() message. what are you guys using? $ LC_ALL=C dpkg -l lilo; man lilo|grep -n -- -H Desired=Unknown/Install/Remove/Purge/Hold | Status=Not/Installed/Config-files/Unpacked/Failed-config/Half-installed |/ Err?=(none)/Hold/Reinst-required/X=both-problems (Status,Err: uppercase=bad) ||/ Name Version Description +++--- ii lilo 1:22.8-3.1 LInux LOader - The Classic OS loader can load Linux and others 108: -H Override fatal halt when a RAID array does not have all disks Greetings Jens -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#442103: initramfs-tools: update-initramfs fails with lilo and a degraded raid1 array
On Sat, Jan 05, 2008 at 08:27:53PM +0100, Jens Thiele wrote: ii lilo 1:22.8-3.1 LInux LOader - The Classic OS loader can load Linux and others 108: -H Override fatal halt when a RAID array does not have all disks Greetings Jens i see, thanks for quick response. as debian supports partial upgrades and initramfs-tools is not gonna start depend on lilo this inocation can only be used after Lenny release. marking as wontfix. -- maks -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#442103: initramfs-tools: update-initramfs fails with lilo and a degraded raid1 array
On 17 Sep 2007, [EMAIL PROTECTED] wrote: hey guys sorry if you still use lilo, you need to do manual intervence, i see no other way. unless a real strong argument comes up, i'll close that bug in a month. AFAIK grub does not support LVM on top of md raid1 whereas lilo does? The installer allows to setup LVM on top of raid and as far as I remember then suggests usage of lilo. Just as explanation / usage scenario: I use lvm on top of raid1 on a laptop and run the raid in degraded state most of the time. From time to time I attach an external usb disk and resync it. (incremental resync that is) IMHO lilo should provide an option in lilo.conf corresponding -H. See also bug: #278373 Greetings Jens -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#442103: initramfs-tools: update-initramfs fails with lilo and a degraded raid1 array
On Thu, 13 Sep 2007, Ionut Ciocirlan wrote: Package: initramfs-tools Version: 0.91a Severity: wishlist When running a degraded raid1 array lilo fails with a fatal error (that can be overriden with -H). Although this is obviously desirable, it effectively breaks any upgrades of packages that run update-initramfs during post-install. Thus my suggestion is running lilo -H and rely on the user to rerun lilo after fixing the array. this is a very specific case: * lilo is not the default since long * if lilo fails you can't boot on powercut so it is urgent for the user to intervene. hey guys sorry if you still use lilo, you need to do manual intervence, i see no other way. unless a real strong argument comes up, i'll close that bug in a month. best regards -- maks -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#442103: initramfs-tools: update-initramfs fails with lilo and a degraded raid1 array
Package: initramfs-tools Version: 0.91a Severity: wishlist When running a degraded raid1 array lilo fails with a fatal error (that can be overriden with -H). Although this is obviously desirable, it effectively breaks any upgrades of packages that run update-initramfs during post-install. Thus my suggestion is running lilo -H and rely on the user to rerun lilo after fixing the array. Ideally, update-initramfs should run lilo -t and grep for this specific problem, and if detected, it would prompt for overriding while warning about the risks, and if the user decides to do so, run lilo -H. -- Package-specific info: -- /proc/cmdline auto BOOT_IMAGE=Linux ro root=fd00 -- /proc/filesystems cramfs reiserfs fuseblk -- lsmod Module Size Used by nvidia 7249460 34 binfmt_misc11400 1 rfcomm 37016 1 l2cap 23232 5 rfcomm button 8208 0 ac 5508 0 battery10308 0 ipv6 239524 16 fuse 42388 1 snd_emu10k1_synth 7296 0 snd_emux_synth 31360 1 snd_emu10k1_synth snd_seq_virmidi 7168 1 snd_emux_synth snd_seq_midi_emul 6208 1 snd_emux_synth snd_emu10k1 123328 3 snd_emu10k1_synth snd_seq_dummy 4036 0 snd_seq_oss29760 0 snd_seq_midi8480 0 snd_seq_midi_event 7360 3 snd_seq_virmidi,snd_seq_oss,snd_seq_midi snd_seq46864 9 snd_emux_synth,snd_seq_virmidi,snd_seq_midi_emul,snd_seq_dummy,snd_seq_oss,snd_seq_midi,snd_seq_midi_event snd_rawmidi23264 3 snd_seq_virmidi,snd_emu10k1,snd_seq_midi firmware_class 9984 1 snd_emu10k1 snd_intel8x0 32412 6 snd_ac97_codec 93220 2 snd_emu10k1,snd_intel8x0 snd_pcm_oss39904 0 ac97_bus2560 1 snd_ac97_codec snd_seq_device 8012 8 snd_emu10k1_synth,snd_emux_synth,snd_emu10k1,snd_seq_dummy,snd_seq_oss,snd_seq_midi,snd_seq,snd_rawmidi snd_mixer_oss 15872 1 snd_pcm_oss snd_pcm72772 7 snd_emu10k1,snd_intel8x0,snd_ac97_codec,snd_pcm_oss snd_timer 21380 5 snd_emu10k1,snd_seq,snd_pcm snd_util_mem4864 2 snd_emux_synth,snd_emu10k1 snd_hwdep 9092 2 snd_emux_synth,snd_emu10k1 snd48804 25 snd_emux_synth,snd_seq_virmidi,snd_emu10k1,snd_seq_oss,snd_seq,snd_rawmidi,snd_intel8x0,snd_ac97_codec,snd_pcm_oss,snd_seq_device,snd_mixer_oss,snd_pcm,snd_timer,snd_hwdep hci_usb16668 2 bluetooth 50148 7 rfcomm,l2cap,hci_usb emu10k1_gp 4032 0 gameport 15304 2 emu10k1_gp pcspkr 3392 0 rtc13144 0 shpchp 31380 0 pci_hotplug29568 1 shpchp soundcore 7840 1 snd snd_page_alloc 9800 3 snd_emu10k1,snd_intel8x0,snd_pcm i2c_nforce2 6144 0 i2c_core 24000 2 nvidia,i2c_nforce2 nvidia_agp 8732 1 agpgart32264 2 nvidia,nvidia_agp tsdev 8320 0 evdev 9664 4 reiserfs 221952 3 dm_mirror 21376 0 dm_snapshot16804 0 dm_mod 52672 11 dm_mirror,dm_snapshot raid1 22784 1 md_mod 74196 3 raid1 ide_cd 36896 0 cdrom 33184 1 ide_cd ide_disk 16832 0 generic 5124 0 [permanent] sd_mod 27584 2 usbhid 26144 0 hid25792 1 usbhid amd74xx13788 0 [permanent] ide_core 114436 4 ide_cd,ide_disk,generic,amd74xx sata_sil 11208 1 firewire_ohci 16960 0 firewire_core 39424 1 firewire_ohci crc_itu_t 2432 1 firewire_core ata_generic 7876 0 libata116912 2 sata_sil,ata_generic scsi_mod 137356 2 sd_mod,libata forcedeth 46088 0 ehci_hcd 31116 0 ohci_hcd 20036 0 usbcore 125704 5 hci_usb,usbhid,ehci_hcd,ohci_hcd thermal13768 0 processor 31368 1 thermal fan 5124 0 -- /etc/kernel-img.conf # Kernel image management overrides # See kernel-img.conf(5) for details do_symlinks = yes relative_links = yes do_bootloader = yes do_bootfloppy = no do_initrd = yes link_in_boot = no -- /etc/initramfs-tools/initramfs.conf MODULES=most BUSYBOX=y KEYMAP=n BOOT=local DEVICE=eth0 NFSROOT=auto -- System Information: Debian Release: lenny/sid APT prefers unstable APT policy: (500, 'unstable'), (500, 'testing') Architecture: i386 (i686) Kernel: Linux 2.6.22-2-k7 (SMP w/1 CPU core) Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/bash Versions of packages
