Re: Boot issue
On Sun, Aug 27, 2023 at 22:51 David Wright wrote: > On Sun 27 Aug 2023 at 14:27:09 (-0500), Tom Browder wrote: > > On Sun, Aug 27, 2023 at 13:27 Greg Wooledge wrote: > > > > > On Sun, Aug 27, 2023 at 08:19:35PM +0200, Hans wrote: > > > > When I boot the system, then the drives are not mounted as set in > > > /etc/fstab. > > > > For example, the SDD should mounted to /space, and the unencrypted > HDD > > > > ... > > > > > Use UUIDs or Labels instead. These won't change, while the device > names > > > *will* change. > > > > > > Please remind me of when they might change. > > As and when the kernel discovers them, ie at boot for those fitted, > and when you plug them is for any others. > > > I'm pretty sure on my latest > > host the debian installer used /dev/sda (and partions 1 and 2) instead > of a > > label or UUID. > > It might be possible to mistakenly read /etc/fstab as showing that, > because of the comment line above the active line: > > # > # / was on /dev/sda4 during installation > UUID=a1b2c3d4-e5f6-1234-dcba-a1b2c3d4e5f6 / ext4 > errors=remount-ro 0 1 > /dev/sr0/media/cdrom0 udf,iso9660 user,noauto 0 0 > > BTW I've not seen the d-i use LABELs, I presume because it can't be > certain that they're always going to be unique. > > > Of course I do want to add drives eventually, so maybe I do need to > change > > to do that safely. > > Some computers can give the user a rude awakening when the kernel > unexpectedly discovers a plugged-in device before the internal drive. Ah, it's been awhile. Yes, I see the comments in /etc/fstab. Thank you for reminding me. Usually I just do "df" and that always shows me the /dev/sdaX so I forget about looking closer. Cheers! -Tom
Re: Boot issue
On Sun 27 Aug 2023 at 14:27:09 (-0500), Tom Browder wrote: > On Sun, Aug 27, 2023 at 13:27 Greg Wooledge wrote: > > > On Sun, Aug 27, 2023 at 08:19:35PM +0200, Hans wrote: > > > When I boot the system, then the drives are not mounted as set in > > /etc/fstab. > > > For example, the SDD should mounted to /space, and the unencrypted HDD > > ... > > > Use UUIDs or Labels instead. These won't change, while the device names > > *will* change. > > > Please remind me of when they might change. As and when the kernel discovers them, ie at boot for those fitted, and when you plug them is for any others. > I'm pretty sure on my latest > host the debian installer used /dev/sda (and partions 1 and 2) instead of a > label or UUID. It might be possible to mistakenly read /etc/fstab as showing that, because of the comment line above the active line: # # / was on /dev/sda4 during installation UUID=a1b2c3d4-e5f6-1234-dcba-a1b2c3d4e5f6 / ext4 errors=remount-ro 0 1 /dev/sr0/media/cdrom0 udf,iso9660 user,noauto 0 0 BTW I've not seen the d-i use LABELs, I presume because it can't be certain that they're always going to be unique. > Of course I do want to add drives eventually, so maybe I do need to change > to do that safely. Some computers can give the user a rude awakening when the kernel unexpectedly discovers a plugged-in device before the internal drive. Cheers, David.
Re: Boot issue
On Sun, Aug 27, 2023 at 13:27 Greg Wooledge wrote: > On Sun, Aug 27, 2023 at 08:19:35PM +0200, Hans wrote: > > When I boot the system, then the drives are not mounted as set in > /etc/fstab. > > For example, the SDD should mounted to /space, and the unencrypted HDD ... > Use UUIDs or Labels instead. These won't change, while the device names > *will* change. Please remind me of when they might change. I'm pretty sure on my latest host the debian installer used /dev/sda (and partions 1 and 2) instead of a label or UUID. Of course I do want to add drives eventually, so maybe I do need to change to do that safely. Thanks. -Tom
[SOLVED] Re: Boot issue
I am answering myself. It must be the opened UUID. Looks like this problem is solved. Thank you all for the fast response! I hope, my question was not too annoying. But I am very very happy, to get this little issue so easily solved - and learned something, too. Thank you all, you made a man happy! Best regards Hans Am Sonntag, 27. August 2023, 21:12:30 CEST schrieb Hans: > So, now I added all UUIDs. But I am not quite sure for the enrcypted /home > partition. The UUID changes when the device is luks-opened. > > Which one must be in the fstab? The one from "lsblk -f /dev/sda4" or > "lsblk -f /dev/mapper/home"? > > Am Sonntag, 27. August 2023, 20:26:46 CEST schrieb Greg Wooledge: > > On Sun, Aug 27, 2023 at 08:19:35PM +0200, Hans wrote: > > > When I boot the system, then the drives are not mounted as set in > > > /etc/fstab. For example, the SDD should mounted to /space, and the > > > unencrypted HDD shall be mounted to /daten. > > > > > > But it seems, sometimes the mountpoints are not correct, so the SDD is > > > not > > > mounted to /space, but to /daten. > > > > This sounds like you used the dynamically assigned device names (e.g. > > /dev/sda1 and so on) in fstab. If that's what you did, stop doing that. > > > > Use UUIDs or Labels instead. These won't change, while the device names > > *will* change.
Re: Boot issue
So, now I added all UUIDs. But I am not quite sure for the enrcypted /home partition. The UUID changes when the device is luks-opened. Which one must be in the fstab? The one from "lsblk -f /dev/sda4" or "lsblk -f /dev/mapper/home"? Am Sonntag, 27. August 2023, 20:26:46 CEST schrieb Greg Wooledge: > On Sun, Aug 27, 2023 at 08:19:35PM +0200, Hans wrote: > > When I boot the system, then the drives are not mounted as set in > > /etc/fstab. For example, the SDD should mounted to /space, and the > > unencrypted HDD shall be mounted to /daten. > > > > But it seems, sometimes the mountpoints are not correct, so the SDD is not > > mounted to /space, but to /daten. > > This sounds like you used the dynamically assigned device names (e.g. > /dev/sda1 and so on) in fstab. If that's what you did, stop doing that. > > Use UUIDs or Labels instead. These won't change, while the device names > *will* change.
Re: Boot issue
Here it is: # /etc/fstab: static file system information. # # proc/proc procdefaults0 0 # /dev/sda3 / ext4defaults,errors=remount-ro 0 1 UUID=819bf8ae-a727-4b5e-97f8-007f58e98f74 /ext4 defaults,errors=remount-ro 0 1 # /dev/sda1 /boot ext2defaults0 2 UUID=069980b1-f1d9-4a11-9e24-147d08b2073f /boot ext2 defaults0 2 # /dev/sda2 noneswapsw 0 0 UUID=2e42eec8-29ed-41d3-8fa6-2869e32605e4 noneswapsw 0 0 /dev/mapper/home /home ext4defaults0 0 # /dev/hdd/media/cdrom0 udf,iso9660 user,noauto 0 0 /dev/cdrom/media/cdrom0 udf,iso9660 user,noauto 0 0 # /dev/fd0/media/floppy0 autorw,user,noauto 0 0 # /dev/sda1 /home/ullhan63/games ext3rw,user,noauto 0 0 # /dev/disk/by-label/UIT-KEY /media vfat noauto,errors=remount-ro0 0 /dev/sdc1 /space ext4defaults0 0 > [...] > > Show us your /etc/fstab. > > Cheers It might be as Greg mentioned: devicename instead of UUID. Hans
Re: Boot issue on system with four SATA devices
On Sun, Aug 27, 2023 at 08:19:35PM +0200, Hans wrote: > Dear list, > > there is a little issue, which I try to solve. On my desptop computer I have > 4 > harddrives: > > SATA 0: HDD 300 GB with Debian + GRUB on MBR (parted in /boot, /, /home > (luks), /var (luks) and /usr (luks) > SATA 1: HDD 300 GB with Win10 > SATA 2: SDD 128 GB (as simple data storage) > SATA 3: HDD 1,5 TB (as simple data storage, splitted as 1TB encrypted and 500 > GB unencrypted) > > Now the issue: > > When I boot the system, then the drives are not mounted as set in /etc/fstab. What is the content of the expected /etc/fstab? > For example, the SDD should mounted to /space, and the unencrypted HDD shall > be mounted to /daten. > > But it seems, sometimes the mountpoints are not correct, so the SDD is not > mounted to /space, but to /daten. > > Also some harddrives are not recognized at all and I then must boot several > times, that all drives and partitions are seen. > > Question one: Is this a problem of the BIOS and motherboard, or a kernel > problem? > > Question two: I remeber to have read, as soon as the kernel hast started (and > it is always starting) the BIOS has no more influence to the hardware. Is > this > correct? If yes, then it is definately a kernel issue. What can I do? Try to understand what is going on ... > I suppose, if this is a hardware issue (I am thinking of, that there is a > timing problem due to speed differences among the slow HDD and the fast SDD), > is there anything I can do (except of replacing the HDD with SDD)? Share with us the content of the /etc/fstab > Any hints welcome! Subject lines that express that some effort has been made. > Best regards > Hans Groeten Geert Stappers -- Silence is hard to parse
Re: Boot issue
Hans composed on 2023-08-27 20:19 (UTC+0200): > When I boot the system, then the drives are not mounted as set in /etc/fstab. > For example, the SDD should mounted to /space, and the unencrypted HDD shall > be mounted to /daten. > But it seems, sometimes the mountpoints are not correct, so the SDD is not > mounted to /space, but to /daten. > Also some harddrives are not recognized at all and I then must boot several > times, that all drives and partitions are seen. Please show output from lsblk -f and content of /etc/fstab. -- Evolution as taught in public schools is, like religion, based on faith, not based on science. Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata
Re: Boot issue
On Sun, Aug 27, 2023 at 08:19:35PM +0200, Hans wrote: > Dear list, > > there is a little issue, which I try to solve. On my desptop computer I have > 4 > harddrives: > > SATA 0: HDD 300 GB with Debian + GRUB on MBR (parted in /boot, /, /home > (luks), /var (luks) and /usr (luks) > SATA 1: HDD 300 GB with Win10 > SATA 2: SDD 128 GB (as simple data storage) > SATA 3: HDD 1,5 TB (as simple data storage, splitted as 1TB encrypted and 500 > GB unencrypted) > > Now the issue: > > When I boot the system, then the drives are not mounted as set in /etc/fstab. > For example, the SDD should mounted to /space, and the unencrypted HDD shall > be mounted to /daten. [...] Show us your /etc/fstab. Cheers -- t signature.asc Description: PGP signature
Re: Boot issue
On Sun, Aug 27, 2023 at 08:19:35PM +0200, Hans wrote: > When I boot the system, then the drives are not mounted as set in /etc/fstab. > For example, the SDD should mounted to /space, and the unencrypted HDD shall > be mounted to /daten. > > But it seems, sometimes the mountpoints are not correct, so the SDD is not > mounted to /space, but to /daten. This sounds like you used the dynamically assigned device names (e.g. /dev/sda1 and so on) in fstab. If that's what you did, stop doing that. Use UUIDs or Labels instead. These won't change, while the device names *will* change.
Re: Boot issue
> When I boot the system, then the drives are not mounted as set in /etc/fstab. I suspect that showing us your `/etc/fstab` would help, Stefan
Re: Boot issue - machine ends up in Busybox
On 16/03/14 02:40 PM, X wrote: Hi, On Wed, I upgraded my debian box (unstable) using aptitude and then shut it down. On booting, it gave me an alert about a disk it couldn't find and brought up the Busybox prompt. I have /boot as a normal partition and other partitions (/, /home, /tmp, /usr, /var, etc.) in a LVM group. On checking the LVM partitions, everything seemed to be fine. Then I did a lvm vgchange -ay and exit, the machine booted up fine after that. I rebooted the machine and had to do the activate the volumegroup again to be able to boot! Why isn't it able to scan the lvm groups and volumes? Is it something to do with systemd? The message before it goes to busybox is shown below if it is helpful. Gave up waiting for root device. Common problems: Boot args (cat /proc/cmdline ) Check root delay = (did the system wait long enough) Check root= (did the system wait for the right device?) Missing modules (cat /proc/modules; ls /dev) ALERT! /dev/disk/by-uuid/66743620-9332-4a1e-a8b1-933eeab1fd89 does not exist. Dropping to shell! modprobe: module ehci-orion not found in modules.dep * More system info: Linux localhost 3.13-1-686-pae #1 SMP Debian 3.13.5-1 (2014-03-04) i686 GNU/Linux $ apt-cache policy lvm2 lvm2: Installed: 2.02.104-2 Candidate: 2.02.104-2 Version table: *** 2.02.104-2 0 500 http://ftp.ca.debian.org/debian/ unstable/main i386 Packages 500 http://ftp.de.debian.org/debian/ unstable/main i386 Packages 100 /var/lib/dpkg/status $ apt-cache policy systemd systemd: Installed: 204-7 Candidate: 204-7 Version table: *** 204-7 0 500 http://ftp.ca.debian.org/debian/ unstable/main i386 Packages 500 http://ftp.de.debian.org/debian/ unstable/main i386 Packages 100 /var/lib/dpkg/status $ sudo /sbin/lvdisplay -c /dev/lvm-gp/glvm01:lvm-gp:3:1:-1:2:3899392:476:-1:0:-1:254:0 /dev/lvm-gp/glvm02:lvm-gp:3:1:-1:1:3899392:476:-1:0:-1:254:1 /dev/lvm-gp/glvm03:lvm-gp:3:1:-1:1:39059456:4768:-1:0:-1:254:2 /dev/lvm-gp/glvm04:lvm-gp:3:1:-1:1:9764864:1192:-1:0:-1:254:3 /dev/lvm-gp/glvm05:lvm-gp:3:1:-1:1:9764864:1192:-1:0:-1:254:4 /dev/lvm-gp/glvm06:lvm-gp:3:1:-1:1:312500224:38147:-1:0:-1:254:5 /dev/lvm-gp/glvm07:lvm-gp:3:1:-1:1:97648640:11920:-1:0:-1:254:6 -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/5325f90a.4090...@fastmail.fm
Re: Boot issue - machine ends up in Busybox
On Sun, 16 Mar 2014 14:40:21 -0400 X list...@fastmail.fm wrote: Hi, On Wed, I upgraded my debian box (unstable) using aptitude and then shut it down. On booting, it gave me an alert about a disk it couldn't find and brought up the Busybox prompt. I have /boot as a normal partition and other partitions (/, /home, /tmp, /usr, /var, etc.) in a LVM group. On checking the LVM partitions, everything seemed to be fine. Then I did a lvm vgchange -ay and exit, the machine booted up fine after that. I rebooted the machine and had to do the activate the volumegroup again to be able to boot! Why isn't it able to scan the lvm groups and volumes? Is it something to do with systemd? The message before it goes to busybox is shown below if it is helpful. Gave up waiting for root device. Common problems: Boot args (cat /proc/cmdline ) Check root delay = (did the system wait long enough) Check root= (did the system wait for the right device?) Missing modules (cat /proc/modules; ls /dev) ALERT! /dev/disk/by-uuid/66743620-9332-4a1e-a8b1-933eeab1fd89 does not exist. Dropping to shell! modprobe: module ehci-orion not found in modules.dep * Been there, done that, reported the bug. I'm surprised how few people this has hit. In my case, the /boot and all the LVM partitions except swap are reiserfs, and in its faulty state, it can see the reiserfs /boot and the LVM swap but not the LVM reiserfs partitions. It appears to be an issue with the new grub-pc-bin 2.02~beta2-7. I downgraded to 2.00-22 along with grub-common, and it booted again. I then downgraded two other 2.02 packages marked as broken, and all was well again. -- Joe -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140316202645.03442...@jretrading.com
Re: Boot issue - machine ends up in Busybox
On 16/03/14 04:26 PM, Joe wrote:
It appears to be an issue with the new grub-pc-bin 2.02~beta2-7. I
downgraded to 2.00-22 along with grub-common, and it booted again. I
then downgraded two other 2.02 packages marked as broken, and all was
well again.
Pinned packages and reinstalled 2.00-22 for
The following NEW packages will be installed:
grub-common{a} grub-pc grub-pc-bin{a} grub2-common{a}
Rebooting...
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/53262fc5.8080...@fastmail.fm
Re: Boot issue - machine ends up in Busybox
On 16/03/14 07:12 PM, KS wrote:
On 16/03/14 04:26 PM, Joe wrote:
It appears to be an issue with the new grub-pc-bin 2.02~beta2-7. I
downgraded to 2.00-22 along with grub-common, and it booted again. I
then downgraded two other 2.02 packages marked as broken, and all was
well again.
Pinned packages and reinstalled 2.00-22 for
The following NEW packages will be installed:
grub-common{a} grub-pc grub-pc-bin{a} grub2-common{a}
Rebooting...
Reboot works normally now. Staying with 2.00-22 for now.
Thanks,
KS
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/5326349e.3080...@fastmail.fm
Re: Boot issue, new installation of Lenny with mirrored /boot and mirrored lvm2
Hi, On Fri, 2009-07-03 at 12:53 +1000, Andrew McGlashan wrote: I have 2x IBM xSeries 346 servers, one has a RAID card (let's call this machine 346a) and the other only has onboard SCSI (346b). [..] it stops at a special initramfs sh after giving up on the lvm volume group (not found) ... from there I can activate the md0 and md1 mirrors and activate the lvm2 logical volumes, I then exit the sh and boot continues as normal. I've checked the kernel and module files for each machine, they are 100% identical. I've checked all files in /boot and /boot/grub and the differences don't give me any clues. Your problem is probably inside your /boot/initrd... file You might want to rebuild it, using update-initramfs Also, did you read /usr/share/doc/mdadm/rootraiddoc.97.html (which is outdated, but gives good overview) Hope this helps, Franklin -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Re: Boot issue, new installation of Lenny with mirrored /boot and mirrored lvm2
Hi Franklin, On Fri, July 3, 2009 4:36 pm, Frank Lin PIAT wrote: Your problem is probably inside your /boot/initrd... file You might want to rebuild it, using update-initramfs Tried that, multiple ways: update-initramfs -u -k all update-initramfs -n -k all Also, did you read /usr/share/doc/mdadm/rootraiddoc.97.html (which is outdated, but gives good overview) No, not yet. But the other 346 machine is working fine and I don't think that I have anything missing in my understanding other than why mdadm isn't assembling the drives as it should on the 'bad' machine when the processes are exactly the same and work fine on the 'good' machine. Hope this helps, Thanks. Kind Regards AndrewM Andrew McGlashan Broadband Solutions now including VoIP -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
