what's this delay thing at boot?
Hi, all -- I have an existing Leap 15.2 system with an nvme SSD, and I'm happy with how it boots. It does not, such that I have ever noticed, exhibit this behavior. However, I have also just built another Leap 15.2 diskfarm:~ # grep . /etc/SUSE-brand /etc/os-release ; uname -a /etc/SUSE-brand:openSUSE /etc/SUSE-brand:VERSION = 15.2 /etc/os-release:NAME="openSUSE Leap" /etc/os-release:VERSION="15.2" /etc/os-release:ID="opensuse-leap" /etc/os-release:ID_LIKE="suse opensuse" /etc/os-release:VERSION_ID="15.2" /etc/os-release:PRETTY_NAME="openSUSE Leap 15.2" /etc/os-release:ANSI_COLOR="0;32" /etc/os-release:CPE_NAME="cpe:/o:opensuse:leap:15.2" /etc/os-release:BUG_REPORT_URL="https://bugs.opensuse.org" /etc/os-release:HOME_URL="https://www.opensuse.org/" Linux diskfarm 5.3.18-lp152.63-default #1 SMP Mon Feb 1 17:31:55 UTC 2021 (98caa86) x86_64 x86_64 x86_64 GNU/Linux system on a single SATA SSD, and as it boots is reports [ slider ] A start job is running for dev-disk-by\x2dlabel-System.device (nns / 1min 30s) that always runs to the 90-second timeout and then fails. What is that, and how do I make it not?!? TIA & HAND :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt
On 15/03/2021 02.38, David T-G wrote:
Hi, all --
I have an existing Leap 15.2 system with an nvme SSD, and I'm happy with how it boots. It does not, such that I have ever noticed, exhibit this behavior.
However, I have also just built another Leap 15.2
diskfarm:~ # grep . /etc/SUSE-brand /etc/os-release ; uname -a /etc/SUSE-brand:openSUSE /etc/SUSE-brand:VERSION = 15.2 /etc/os-release:NAME="openSUSE Leap" /etc/os-release:VERSION="15.2" /etc/os-release:ID="opensuse-leap" /etc/os-release:ID_LIKE="suse opensuse" /etc/os-release:VERSION_ID="15.2" /etc/os-release:PRETTY_NAME="openSUSE Leap 15.2" /etc/os-release:ANSI_COLOR="0;32" /etc/os-release:CPE_NAME="cpe:/o:opensuse:leap:15.2" /etc/os-release:BUG_REPORT_URL="https://bugs.opensuse.org" /etc/os-release:HOME_URL="https://www.opensuse.org/" Linux diskfarm 5.3.18-lp152.63-default #1 SMP Mon Feb 1 17:31:55 UTC 2021 (98caa86) x86_64 x86_64 x86_64 GNU/Linux
system on a single SATA SSD, and as it boots is reports
[ slider ] A start job is running for dev-disk-by\x2dlabel-System.device (nns / 1min 30s)
that always runs to the 90-second timeout and then fails. What is that, and how do I make it not?!? The behaviour is just that message, or is there a problem?
-- Cheers / Saludos, Carlos E. R. (from 15.2 x86_64 at Telcontar)
Carlos, et al -- ...and then Carlos E. R. said... % % On 15/03/2021 02.38, David T-G wrote: % > ... % >system on a single SATA SSD, and as it boots is reports % > % > [ slider ] A start job is running for dev-disk-by\x2dlabel-System.device (nns / 1min 30s) % > % >that always runs to the 90-second timeout and then fails. What is that, % >and how do I make it not?!? % The behaviour is just that message, or is there a problem? It reports a failure, but I can't for the life of me figure out what it's trying to do or discover, and afterwards the system continues and boots happily. It's just looking for something, waiting a good long time, and then giving up happily when it doesn't find it. Thanks again & HAND :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt
On 15/03/2021 11.06, David T-G wrote:
Carlos, et al --
...and then Carlos E. R. said... % % On 15/03/2021 02.38, David T-G wrote: % > ... % >system on a single SATA SSD, and as it boots is reports % > % > [ slider ] A start job is running for dev-disk-by\x2dlabel-System.device (nns / 1min 30s) % > % >that always runs to the 90-second timeout and then fails. What is that, % >and how do I make it not?!? % The behaviour is just that message, or is there a problem?
It reports a failure, but I can't for the life of me figure out what it's trying to do or discover, and afterwards the system continues and boots happily. It's just looking for something, waiting a good long time, and then giving up happily when it doesn't find it.
Ah, ok, it is a delay and then boots succeeds. It does succeed? Is that the exact text? I would guess you have a mount defined on fstab that can not be mounted. -- Cheers / Saludos, Carlos E. R. (from 15.2 x86_64 at Telcontar)
Carlos, et al -- ...and then Carlos E. R. said... % % On 15/03/2021 11.06, David T-G wrote: % > % >...and then Carlos E. R. said... % >% % >% The behaviour is just that message, or is there a problem? % > % >It reports a failure, but I can't for the life of me figure out what it's ... % >then giving up happily when it doesn't find it. % % Ah, ok, it is a delay and then boots succeeds. It does succeed? Yep. She just moves right along :-/ % % Is that the exact text? https://paste.opensuse.org/16547964 % % I would guess you have a mount defined on fstab that can not be mounted. This has happened since the first boot with just the local SSD on board and only / swap /mnt/ssd in the fstab file :-( % % -- % Cheers / Saludos, % % Carlos E. R. % (from 15.2 x86_64 at Telcontar) % Thanks again & HANN :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt
On 16/03/2021 02.20, David T-G wrote:
Carlos, et al --
...and then Carlos E. R. said... % % On 15/03/2021 11.06, David T-G wrote: % > % >...and then Carlos E. R. said... % >% % >% The behaviour is just that message, or is there a problem? % > % >It reports a failure, but I can't for the life of me figure out what it's ... % >then giving up happily when it doesn't find it. % % Ah, ok, it is a delay and then boots succeeds. It does succeed?
Yep. She just moves right along :-/
% % Is that the exact text?
I can not guess what is the actual label. :-/ Or what systemd service to try find. Some systemd parlance.
% % I would guess you have a mount defined on fstab that can not be mounted.
This has happened since the first boot with just the local SSD on board and only
/ swap /mnt/ssd
in the fstab file :-(
-- Cheers / Saludos, Carlos E. R. (from 15.2 x86_64 at Telcontar)
On 2021-03-15 20:20:44 David T-G wrote:
|% |% I would guess you have a mount defined on fstab that can not be mounted. | |This has happened since the first boot with just the local SSD on board |and only | | / | swap | /mnt/ssd | |in the fstab file :-(
If you have something like the NFS (or samba?) client installed and it looks for a remote file server that isn't responding, that will cause a delay. Leslie -- openSUSE Leap 15.2 x86_64
On 16/03/2021 04.09, J Leslie Turriff wrote:
On 2021-03-15 20:20:44 David T-G wrote:
|% |% I would guess you have a mount defined on fstab that can not be mounted. | |This has happened since the first boot with just the local SSD on board |and only | | / | swap | /mnt/ssd | |in the fstab file :-(
If you have something like the NFS (or samba?) client installed and it looks for a remote file server that isn't responding, that will cause a delay.
Or an encrypted partition that wants a password, then times out. Possibly an automount entry, too. -- Cheers / Saludos, Carlos E. R. (from 15.2 x86_64 at Telcontar)
Leslie & Carlos, et al -- ...and then Carlos E. R. said... % % On 16/03/2021 04.09, J Leslie Turriff wrote: % >On 2021-03-15 20:20:44 David T-G wrote: % >>| % >>|This has happened since the first boot with just the local SSD on board % >>|and only % >>| % >>| / % >>| swap % >>| /mnt/ssd % >>| % >>|in the fstab file :-( % > % > If you have something like the NFS (or samba?) client installed and it looks for a remote % >file server that isn't responding, that will cause a delay. I could imagine that, except that we don't have that at an initial build with nothing else listed in fstab. % % Or an encrypted partition that wants a password, then times out. That's a good point, although I don't have any of those, so I can rule out that option via local knowledge. But the next guy might have that. % % Possibly an automount entry, too. Aha! True; those don't show up in fstab, do they? Local knowledge for me again, but a good place to check. % % -- % Cheers / Saludos, % % Carlos E. R. % (from 15.2 x86_64 at Telcontar) % Thanks again & HAND :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt
On 18/03/2021 12.27, David T-G wrote:
Leslie & Carlos, et al --
...and then Carlos E. R. said... % % On 16/03/2021 04.09, J Leslie Turriff wrote: % >On 2021-03-15 20:20:44 David T-G wrote: % >>| % >>|This has happened since the first boot with just the local SSD on board % >>|and only % >>| % >>| / % >>| swap % >>| /mnt/ssd % >>| % >>|in the fstab file :-( % > % > If you have something like the NFS (or samba?) client installed and it looks for a remote % >file server that isn't responding, that will cause a delay.
I could imagine that, except that we don't have that at an initial build with nothing else listed in fstab.
% % Or an encrypted partition that wants a password, then times out.
That's a good point, although I don't have any of those, so I can rule out that option via local knowledge. But the next guy might have that.
% % Possibly an automount entry, too.
Aha! True; those don't show up in fstab, do they? Local knowledge for me again, but a good place to check.
They show or not depending on the method used, I think. -- Cheers / Saludos, Carlos E. R. (from 15.2 x86_64 at Telcontar)
On 2021-03-16 06:42:48 Carlos E. R. wrote:
|On 16/03/2021 04.09, J Leslie Turriff wrote: |> On 2021-03-15 20:20:44 David T-G wrote: |>> |% |>> |% I would guess you have a mount defined on fstab that can not be |>> | mounted. |>> | |>> |This has happened since the first boot with just the local SSD on |>> | board and only |>> | |>> | / |>> | swap |>> | /mnt/ssd |>> | |>> |in the fstab file :-( |> |> If you have something like the NFS (or samba?) client installed and it |> looks for a remote file server that isn't responding, that will cause a |> delay. | |Or an encrypted partition that wants a password, then times out. | |Possibly an automount entry, too.
In my case, during bootup my machine stops at the message | Reached target Remote File Systems which I assume is NFS, Samba, etc. Could this also be USB-connected drives? Leslie -- openSUSE Leap 15.2 x86_64
On 19/03/2021 07.11, J Leslie Turriff wrote:
On 2021-03-16 06:42:48 Carlos E. R. wrote:
|On 16/03/2021 04.09, J Leslie Turriff wrote: |> On 2021-03-15 20:20:44 David T-G wrote: |>> |% |>> |% I would guess you have a mount defined on fstab that can not be |>> | mounted. |>> | |>> |This has happened since the first boot with just the local SSD on |>> | board and only |>> | |>> | / |>> | swap |>> | /mnt/ssd |>> | |>> |in the fstab file :-( |> |> If you have something like the NFS (or samba?) client installed and it |> looks for a remote file server that isn't responding, that will cause a |> delay. | |Or an encrypted partition that wants a password, then times out. | |Possibly an automount entry, too.
In my case, during bootup my machine stops at the message | Reached target Remote File Systems which I assume is NFS, Samba, etc. Could this also be USB-connected drives?
It is unlikely to require an usb mount at boot, but it is possible. Although... I have a machine that does mount some external disks (encrypted) via usb at boot, but they are referred by label or id in crypttab/fstab. One of them I consider optional, and when it is off it does cause a delay. I suspect a bug in crypttab. But the system is not aware of the disk being on USB. Perhaps the "magic" that finds the partitions (udev?) -- Cheers / Saludos, Carlos E. R. (from 15.2 x86_64 at Telcontar)
On 2021-03-19 05:23:47 Carlos E. R. wrote:
|> In my case, during bootup my machine stops at the message |> |> | Reached target Remote File Systems |> |> which I assume is NFS, Samba, etc. Could this also be USB-connected |> drives? | |It is unlikely to require an usb mount at boot, but it is possible. | |Although... | |I have a machine that does mount some external disks (encrypted) via usb |at boot, but they are referred by label or id in crypttab/fstab. One of |them I consider optional, and when it is off it does cause a delay. I |suspect a bug in crypttab. | |But the system is not aware of the disk being on USB. Perhaps the |"magic" that finds the partitions (udev?)
I'm just wondering what the definition of "Remote File Systems" is for the person who created that message. :-) Leslie -- openSUSE Leap 15.2 x86_64
On 20/03/2021 04.22, J Leslie Turriff wrote:
On 2021-03-19 05:23:47 Carlos E. R. wrote:
|> In my case, during bootup my machine stops at the message |> |> | Reached target Remote File Systems |> |> which I assume is NFS, Samba, etc. Could this also be USB-connected |> drives? | |It is unlikely to require an usb mount at boot, but it is possible. | |Although... | |I have a machine that does mount some external disks (encrypted) via usb |at boot, but they are referred by label or id in crypttab/fstab. One of |them I consider optional, and when it is off it does cause a delay. I |suspect a bug in crypttab. | |But the system is not aware of the disk being on USB. Perhaps the |"magic" that finds the partitions (udev?)
I'm just wondering what the definition of "Remote File Systems" is for the person who created that message. :-)
NFS and SMB, I think. -- Cheers / Saludos, Carlos E. R. (from 15.2 x86_64 at Telcontar)
On 2021-03-20 05:03:03 Carlos E. R. wrote:
|On 20/03/2021 04.22, J Leslie Turriff wrote: |> On 2021-03-19 05:23:47 Carlos E. R. wrote: |>> |> In my case, during bootup my machine stops at the message |>> |> |>> |> | Reached target Remote File Systems |>> |> |>> |> which I assume is NFS, Samba, etc. Could this also be USB-connected |>> |> drives? |>> | |>> |It is unlikely to require an usb mount at boot, but it is possible. |>> | |>> |Although... |>> | |>> |I have a machine that does mount some external disks (encrypted) via |>> | usb at boot, but they are referred by label or id in crypttab/fstab. |>> | One of them I consider optional, and when it is off it does cause a |>> | delay. I suspect a bug in crypttab. |>> | |>> |But the system is not aware of the disk being on USB. Perhaps the |>> |"magic" that finds the partitions (udev?) |> |> I'm just wondering what the definition of "Remote File Systems" is for |> the person who created that message. :-) | |NFS and SMB, I think.
One might suppose that USB-connected drives could also be classed as remote, since they're not guaranteed to be accessible at boot-time. But in the absence of a real definition, it's all speculative. :-) Leslie -- openSUSE Leap 15.2 x86_64
On 24/03/2021 04.40, J Leslie Turriff wrote:
On 2021-03-20 05:03:03 Carlos E. R. wrote:
|On 20/03/2021 04.22, J Leslie Turriff wrote: |> On 2021-03-19 05:23:47 Carlos E. R. wrote: |>> |> In my case, during bootup my machine stops at the message |>> |> |>> |> | Reached target Remote File Systems
...
|> I'm just wondering what the definition of "Remote File Systems" is for |> the person who created that message. :-) | |NFS and SMB, I think.
One might suppose that USB-connected drives could also be classed as remote, since they're not guaranteed to be accessible at boot-time. But in the absence of a real definition, it's all speculative. :-)
USB has another name that is eluding me this instant... plugable? An USB disk is directly connected to this computer. A remote filesystem is connected to another computer. -- Cheers / Saludos, Carlos E. R. (from 15.2 x86_64 at Telcontar)
On Tue, 23 Mar 2021 22:40:19 -0500 J Leslie Turriff <jlturriff@mail.com> wrote:
On 2021-03-20 05:03:03 Carlos E. R. wrote:
|On 20/03/2021 04.22, J Leslie Turriff wrote: |> On 2021-03-19 05:23:47 Carlos E. R. wrote: |>> |> In my case, during bootup my machine stops at the message |>> |> |>> |> | Reached target Remote File Systems |>> |> |>> |> which I assume is NFS, Samba, etc. Could this also be USB-connected |>> |> drives? |>> | |>> |It is unlikely to require an usb mount at boot, but it is possible. |>> | |>> |Although... |>> | |>> |I have a machine that does mount some external disks (encrypted) via |>> | usb at boot, but they are referred by label or id in crypttab/fstab. |>> | One of them I consider optional, and when it is off it does cause a |>> | delay. I suspect a bug in crypttab. |>> | |>> |But the system is not aware of the disk being on USB. Perhaps the |>> |"magic" that finds the partitions (udev?) |> |> I'm just wondering what the definition of "Remote File Systems" is for |> the person who created that message. :-) | |NFS and SMB, I think.
One might suppose that USB-connected drives could also be classed as remote, since they're not guaranteed to be accessible at boot-time. But in the absence of a real definition, it's all speculative. :-)
Remote doesn't have anything to do with availability at boot, IMHO anyway. It means 'connected over the/a network' 'connected to another host'.
Leslie
On 3/14/21 8:38 PM, David T-G wrote:
system on a single SATA SSD, and as it boots is reports
[ slider ] A start job is running for dev-disk-by\x2dlabel-System.device (nns / 1min 30s)
that always runs to the 90-second timeout and then fails. What is that, and how do I make it not?!?
In addition to what others have said, you should also check "/etc/default/grub". Perhaps use: grep resume /etc/default/grub The "resume=" parameter is where to find a possible hibernation image.
Neil, et al -- ...and then Neil Rickert said... % % In addition to what others have said, you should also check % "/etc/default/grub". Perhaps use: % % grep resume /etc/default/grub OOOOOH! We have a bingo: diskfarm:~ # grep resume /etc/default/grub GRUB_CMDLINE_LINUX_DEFAULT="splash=silent resume=/dev/disk/by-label/System Reserved mitigations=auto quiet" I see why I don't see that on my desktop, too: davidtg@gezebel:~> grep resume /etc/default/grub ### GRUB_CMDLINE_LINUX_DEFAULT="splash=silent resume=/dev/disk/by-label/gezebel-swap mitigations=auto quiet" GRUB_CMDLINE_LINUX_DEFAULT="resume=/dev/disk/by-label/gezebel-swap mitigations=auto quiet" Apparently I forgot to write down that change: davidtg@gezebel:~> grep resume /mnt/data/OS/OS-notes.gezebel.txt davidtg@gezebel:~> I can fix that :-) % % The "resume=" parameter is where to find a possible hibernation image. Makes sense ... now :-/ Since diskfarm is a "server" and gezebel is a desktop and neither ever gets shut down to pause, we don't need to go there. THANKS!!! :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt
On 18/03/2021 12.23, David T-G wrote:
Neil, et al --
...and then Neil Rickert said... % % In addition to what others have said, you should also check % "/etc/default/grub". Perhaps use: % % grep resume /etc/default/grub
OOOOOH! We have a bingo:
diskfarm:~ # grep resume /etc/default/grub GRUB_CMDLINE_LINUX_DEFAULT="splash=silent resume=/dev/disk/by-label/System Reserved mitigations=auto quiet"
Mystery explained :-) ...
% The "resume=" parameter is where to find a possible hibernation image.
Makes sense ... now :-/ Since diskfarm is a "server" and gezebel is a desktop and neither ever gets shut down to pause, we don't need to go there.
Arguably, hibernation can be used instead of poweroff triggered by a power failure (UPS monitoring daemon). -- Cheers / Saludos, Carlos E. R. (from 15.2 x86_64 at Telcontar)
Carlos, et al -- ...and then Carlos E. R. said... % % On 18/03/2021 12.23, David T-G wrote: ... % >Makes sense ... now :-/ Since diskfarm is a "server" and gezebel is a % >desktop and neither ever gets shut down to pause, we don't need to go % >there. % % Arguably, hibernation can be used instead of poweroff triggered by a % power failure (UPS monitoring daemon). Hmmmmm ... Good point. Maybe I'll play with configuring hibernation one day. Good thing I always define "enough" swap space :-) Thanks again & HAND (um, que tenga un buen dia, no?) :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt
Neil, et al -- ...and then David T-G home said... % % ...and then Neil Rickert said... % % % % In addition to what others have said, you should also check % % "/etc/default/grub". Perhaps use: % % % % grep resume /etc/default/grub % ... % I see why I don't see that on my desktop, too: % % davidtg@gezebel:~> grep resume /etc/default/grub % ### GRUB_CMDLINE_LINUX_DEFAULT="splash=silent resume=/dev/disk/by-label/gezebel-swap mitigations=auto quiet" % GRUB_CMDLINE_LINUX_DEFAULT="resume=/dev/disk/by-label/gezebel-swap mitigations=auto quiet" [snip] No, I *don't* see, because I got rid of the splash rather than the resume. No wonder I didn't write that down ... The first thing I'll do is get rid of resume= and see how it goes. If successful, that will leave me scratching my head as to why gezebel DOES look but DOESN'T take 90 seconds, but I won't scratch for too long. If not, then I'm no worse off than I am now. So expect an update ... :-) PS -- here's another place where I'd rather not be writing in /etc/default but instead to a user-controlled dir that is intended for local prefs :-/ HAND :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt
On 3/18/21 6:23 AM, David T-G wrote:
% grep resume /etc/default/grub
OOOOOH! We have a bingo:
diskfarm:~ # grep resume /etc/default/grub GRUB_CMDLINE_LINUX_DEFAULT="splash=silent resume=/dev/disk/by-label/System Reserved mitigations=auto quiet"
This needs a bug report. Apparently the label for that partition is "System Reserved" which contains a space. So quoting is needed. It should have used GRUB_CMDLINE_LINUX_DEFAULT="splash=silent resume=/dev/disk/by-label/System\ Reserved mitigations=auto quiet" or GRUB_CMDLINE_LINUX_DEFAULT="splash=silent resume=/dev/disk/by-label/'System Reserved' mitigations=auto quiet" or something similar. Or it should ignore your settings and use UUID for that particular partition.
On 18/03/2021 17.12, Neil Rickert wrote:
On 3/18/21 6:23 AM, David T-G wrote:
% grep resume /etc/default/grub
OOOOOH! We have a bingo:
diskfarm:~ # grep resume /etc/default/grub GRUB_CMDLINE_LINUX_DEFAULT="splash=silent resume=/dev/disk/by-label/System Reserved mitigations=auto quiet"
This needs a bug report.
You are right, I missed that lone "Reserved".
Apparently the label for that partition is "System Reserved" which contains a space. So quoting is needed. It should have used
GRUB_CMDLINE_LINUX_DEFAULT="splash=silent resume=/dev/disk/by-label/System\ Reserved mitigations=auto quiet"
or
GRUB_CMDLINE_LINUX_DEFAULT="splash=silent resume=/dev/disk/by-label/'System Reserved' mitigations=auto quiet"
or something similar. Or it should ignore your settings and use UUID for that particular partition.
I understood that spaces were prohibited in partition labels. I don't know how to find this out for sure, but I see the labels I use, and I see I used "_" instead of spaces. -- Cheers / Saludos, Carlos E. R. (from 15.2 x86_64 at Telcontar)
On Thu, 18 Mar 2021 19:48:55 +0100 "Carlos E. R." <robin.listas@telefonica.net> wrote:
On 18/03/2021 17.12, Neil Rickert wrote:
On 3/18/21 6:23 AM, David T-G wrote:
% grep resume /etc/default/grub
OOOOOH! We have a bingo:
diskfarm:~ # grep resume /etc/default/grub GRUB_CMDLINE_LINUX_DEFAULT="splash=silent resume=/dev/disk/by-label/System Reserved mitigations=auto quiet"
This needs a bug report.
You are right, I missed that lone "Reserved".
Apparently the label for that partition is "System Reserved" which contains a space. So quoting is needed. It should have used
GRUB_CMDLINE_LINUX_DEFAULT="splash=silent resume=/dev/disk/by-label/System\ Reserved mitigations=auto quiet"
or
GRUB_CMDLINE_LINUX_DEFAULT="splash=silent resume=/dev/disk/by-label/'System Reserved' mitigations=auto quiet"
or something similar. Or it should ignore your settings and use UUID for that particular partition.
I understood that spaces were prohibited in partition labels. I don't know how to find this out for sure, but I see the labels I use, and I see I used "_" instead of spaces.
They are definitely not prohibited in partition labels. I have for example: /dev/sda1: LABEL="ESP" UUID="A2D4-EB6B" TYPE="vfat" PARTLABEL="EFI system partition" PARTUUID="2d6563ec-fd02-4309-bc44-3b5343fc88f2" (sorry for the line wrap) But that's irrelevant because by-label uses the filesystem label (a.k.a. volume label). If you want to use the partition label then you need to use by-partlabel. This https://wiki.archlinux.org/index.php/persistent_block_device_naming archlinux page has examples showing filesystem labels with spaces as well, but I don't know if they actually work. The maximum lengths of filesystem labels vary depending on the filesystem.
On 18/03/2021 21.24, Dave Howorth wrote:
On Thu, 18 Mar 2021 19:48:55 +0100 "Carlos E. R." <> wrote:
On 18/03/2021 17.12, Neil Rickert wrote:
On 3/18/21 6:23 AM, David T-G wrote:
% grep resume /etc/default/grub
OOOOOH! We have a bingo:
diskfarm:~ # grep resume /etc/default/grub GRUB_CMDLINE_LINUX_DEFAULT="splash=silent resume=/dev/disk/by-label/System Reserved mitigations=auto quiet"
This needs a bug report.
You are right, I missed that lone "Reserved".
Apparently the label for that partition is "System Reserved" which contains a space. So quoting is needed. It should have used
GRUB_CMDLINE_LINUX_DEFAULT="splash=silent resume=/dev/disk/by-label/System\ Reserved mitigations=auto quiet"
or
GRUB_CMDLINE_LINUX_DEFAULT="splash=silent resume=/dev/disk/by-label/'System Reserved' mitigations=auto quiet"
or something similar. Or it should ignore your settings and use UUID for that particular partition.
I understood that spaces were prohibited in partition labels. I don't know how to find this out for sure, but I see the labels I use, and I see I used "_" instead of spaces.
They are definitely not prohibited in partition labels. I have for example:
/dev/sda1: LABEL="ESP" UUID="A2D4-EB6B" TYPE="vfat" PARTLABEL="EFI system partition" PARTUUID="2d6563ec-fd02-4309-bc44-3b5343fc88f2"
(sorry for the line wrap)
But that's irrelevant because by-label uses the filesystem label (a.k.a. volume label). If you want to use the partition label then you need to use by-partlabel.
This https://wiki.archlinux.org/index.php/persistent_block_device_naming archlinux page has examples showing filesystem labels with spaces as well, but I don't know if they actually work. The maximum lengths of filesystem labels vary depending on the filesystem.
Definitely a bug report is in order. Find out if spaces are allowed or not on filesystem labels; if not, then find out what wrote that space, and add detection in YaST for the situation and then demand the user to change the label. And if spaces are allowed, handle them. -- Cheers / Saludos, Carlos E. R. (from 15.2 x86_64 at Telcontar)
Carlos, et al -- ...and then Carlos E. R. said... % % On 18/03/2021 17.12, Neil Rickert wrote: % >On 3/18/21 6:23 AM, David T-G wrote: ... % >> diskfarm:~ # grep resume /etc/default/grub % >> GRUB_CMDLINE_LINUX_DEFAULT="splash=silent resume=/dev/disk/by-label/System Reserved mitigations=auto quiet" % > ... % >Apparently the label for that partition is "System Reserved" which % >contains a space. So quoting is needed. It should have used ... % % I understood that spaces were prohibited in partition labels. I % don't know how to find this out for sure, but I see the labels I % use, and I see I used "_" instead of spaces. How does a swap partition (swap volume?) have a volume label, though?!? Now that I know it's there, that's what has me confused. It is NOT shown as a partition label, and I don't see how swap should have a volume label. So what is the thing?!? Any hints on how I dig into this one? TIA again & HANN :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt
David T-G composed on 2021-03-21 00:48 (UTC-0400):
How does a swap partition (swap volume?) have a volume label, though?!?
mkswap -L swapperlabel /dev/sd.... -- Evolution as taught in public schools, like religion, is based on faith, not on science. Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/
Felix, et al -- ...and then Felix Miata said... % % David T-G composed on 2021-03-21 00:48 (UTC-0400): % % > How does a swap partition (swap volume?) have a volume label, though?!? % % mkswap -L swapperlabel /dev/sd.... Well. Um. That was simple :-) Thanks! Now ... How can I *change* a swap label? <poke> Ahhhh... Thank you, man(1) with the -k flag! diskfarm:~ # swaplabel -L diskfarm-swap /dev/sda1 diskfarm:~ # swaplabel /dev/sda1 LABEL: diskfarm-swap UUID: 3982f3a2-3e59-4658-b18d-5a036572c6c7 Ta-daaaaa! We'll probably need another grub2-mkconfig run, but we'll have a look at the next boot. Thanks again & HANN (for real this time; the alarm goes off in five hours!) :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt
On Sun, 21 Mar 2021 01:07:09 -0400 David T-G <davidtg-robot@justpickone.org> wrote:
Felix, et al --
...and then Felix Miata said... % % David T-G composed on 2021-03-21 00:48 (UTC-0400): % % > How does a swap partition (swap volume?) have a volume label, though?!? % % mkswap -L swapperlabel /dev/sd....
Well. Um. That was simple :-) Thanks!
Now ... How can I *change* a swap label? <poke> Ahhhh... Thank you, man(1) with the -k flag!
diskfarm:~ # swaplabel -L diskfarm-swap /dev/sda1 diskfarm:~ # swaplabel /dev/sda1 LABEL: diskfarm-swap UUID: 3982f3a2-3e59-4658-b18d-5a036572c6c7
Ta-daaaaa!
Indeed, and for those curious about spaces in names: # swaplabel /dev/sda7 UUID: 964c2f38-5789-4fe6-b6e5-65c0b486ee7c # swaplabel -L 'swap partition' /dev/sda7 # swaplabel /dev/sda7 LABEL: swap partition UUID: 964c2f38-5789-4fe6-b6e5-65c0b486ee7c # ls -goh /dev/disk/by-label/ | grep swap lrwxrwxrwx 1 10 Mar 21 09:11 swap\x20partition -> ../../sda7 # So I think it is possible to have spaces in volume labels and the /dev/disk/by-label/ hierarchy encodes them.
We'll probably need another grub2-mkconfig run, but we'll have a look at the next boot.
Thanks again & HANN (for real this time; the alarm goes off in five hours!)
:-D
On 21/03/2021 10.19, Dave Howorth wrote:
On Sun, 21 Mar 2021 01:07:09 -0400 David T-G <davidtg-robot@justpickone.org> wrote:
Felix, et al --
...and then Felix Miata said... % % David T-G composed on 2021-03-21 00:48 (UTC-0400): % % > How does a swap partition (swap volume?) have a volume label, though?!? % % mkswap -L swapperlabel /dev/sd....
Well. Um. That was simple :-) Thanks!
Now ... How can I *change* a swap label? <poke> Ahhhh... Thank you, man(1) with the -k flag!
diskfarm:~ # swaplabel -L diskfarm-swap /dev/sda1 diskfarm:~ # swaplabel /dev/sda1 LABEL: diskfarm-swap UUID: 3982f3a2-3e59-4658-b18d-5a036572c6c7
Ta-daaaaa!
Indeed, and for those curious about spaces in names:
# swaplabel /dev/sda7 UUID: 964c2f38-5789-4fe6-b6e5-65c0b486ee7c # swaplabel -L 'swap partition' /dev/sda7 # swaplabel /dev/sda7 LABEL: swap partition UUID: 964c2f38-5789-4fe6-b6e5-65c0b486ee7c # ls -goh /dev/disk/by-label/ | grep swap lrwxrwxrwx 1 10 Mar 21 09:11 swap\x20partition -> ../../sda7 #
So I think it is possible to have spaces in volume labels and the /dev/disk/by-label/ hierarchy encodes them.
Ok, so a bug report is missing: YaST must handle the situation, in whatever direction. David, it's on you to write it ;-) Like... "somehow (unknown how) a partition was labelled with a space and yast did not handle it. No error printed, no correct handling when writing grub" etc etc. -- Cheers / Saludos, Carlos E. R. (from 15.2 x86_64 at Telcontar)
Carlos, et al -- ...and then Carlos E. R. said... % % Ok, so a bug report is missing: YaST must handle the situation, in % whatever direction. David, it's on you to write it ;-) Fine by me :-) % % Like... "somehow (unknown how) a partition was labelled with a space % and yast did not handle it. No error printed, no correct handling % when writing grub" etc etc. OK; that's fair :-) That isn't quite what I meant, though. Any random bugzilla? Is there a SuSE 'zilla? And, just maybe, how would you categorize this one? HAND :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt
On 21/03/2021 15.12, David T-G wrote:
Carlos, et al --
...and then Carlos E. R. said... % % Ok, so a bug report is missing: YaST must handle the situation, in % whatever direction. David, it's on you to write it ;-)
Fine by me :-)
% % Like... "somehow (unknown how) a partition was labelled with a space % and yast did not handle it. No error printed, no correct handling % when writing grub" etc etc.
OK; that's fair :-) That isn't quite what I meant, though.
Any random bugzilla? Is there a SuSE 'zilla? And, just maybe, how would you categorize this one?
<https://en.opensuse.org/openSUSE:Submitting_bug_reports> I would categorize it as "installation". -- Cheers / Saludos, Carlos E. R. (from 15.2 x86_64 at Telcontar)
On 21/03/2021 05.48, David T-G wrote:
Carlos, et al --
...and then Carlos E. R. said... % % On 18/03/2021 17.12, Neil Rickert wrote: % >On 3/18/21 6:23 AM, David T-G wrote: ... % >> diskfarm:~ # grep resume /etc/default/grub % >> GRUB_CMDLINE_LINUX_DEFAULT="splash=silent resume=/dev/disk/by-label/System Reserved mitigations=auto quiet" % > ... % >Apparently the label for that partition is "System Reserved" which % >contains a space. So quoting is needed. It should have used ... % % I understood that spaces were prohibited in partition labels. I % don't know how to find this out for sure, but I see the labels I % use, and I see I used "_" instead of spaces.
How does a swap partition (swap volume?) have a volume label, though?!? Now that I know it's there, that's what has me confused. It is NOT shown as a partition label, and I don't see how swap should have a volume label. So what is the thing?!?
Any hints on how I dig into this one?
You write a label on Swap the same you do it on any filesystem: MKSWAP(8) System Administration MKSWAP(8) NAME mkswap - set up a Linux swap area ... -L, --label label Specify a label for the device, to allow swapon by label. -- Cheers / Saludos, Carlos E. R. (from 15.2 x86_64 at Telcontar)
Neil, et al -- ...and then Neil Rickert said... % % On 3/18/21 6:23 AM, David T-G wrote: % > % grep resume /etc/default/grub % > % > OOOOOH! We have a bingo: % > % > diskfarm:~ # grep resume /etc/default/grub % > GRUB_CMDLINE_LINUX_DEFAULT="splash=silent resume=/dev/disk/by-label/System Reserved mitigations=auto quiet" % % This needs a bug report. % % Apparently the label for that partition is "System Reserved" which % contains a space. So quoting is needed. It should have used [snip] Actually, maybe not. When I look via ls, I see an encoded space: diskfarm:~ # ls -goh /dev/disk/by-label/ | grep Reserved lrwxrwxrwx 1 10 Mar 13 03:55 System\x20Reserved -> ../../sda1 diskfarm:~ # ls -goh /dev/disk/by*/* | grep sda1 lrwxrwxrwx 1 10 Mar 13 03:55 /dev/disk/by-id/ata-SanDisk_SD6SB1M128G1001_142000400317-part1 -> ../../sda1 lrwxrwxrwx 1 10 Mar 13 03:55 /dev/disk/by-id/scsi-0ATA_SanDisk_SD6SB1M1_142000400317-part1 -> ../../sda1 lrwxrwxrwx 1 10 Mar 13 03:55 /dev/disk/by-id/scsi-1ATA_SanDisk_SD6SB1M128G1001_142000400317-part1 -> ../../sda1 lrwxrwxrwx 1 10 Mar 13 03:55 /dev/disk/by-id/scsi-35001b44c1b7723bd-part1 -> ../../sda1 lrwxrwxrwx 1 10 Mar 13 03:55 /dev/disk/by-id/scsi-SATA_SanDisk_SD6SB1M1_142000400317-part1 -> ../../sda1 lrwxrwxrwx 1 10 Mar 13 03:55 /dev/disk/by-id/wwn-0x5001b44c1b7723bd-part1 -> ../../sda1 lrwxrwxrwx 1 10 Mar 13 03:55 /dev/disk/by-label/System\x20Reserved -> ../../sda1 lrwxrwxrwx 1 10 Mar 13 03:55 /dev/disk/by-partlabel/diskfarm-swap -> ../../sda1 lrwxrwxrwx 1 10 Mar 13 03:55 /dev/disk/by-partuuid/9b848a25-dca4-4f4e-a5e4-c5808c71b9d7 -> ../../sda1 lrwxrwxrwx 1 10 Mar 13 03:55 /dev/disk/by-path/pci-0000:00:1f.2-ata-1-part1 -> ../../sda1 lrwxrwxrwx 1 10 Mar 13 03:55 /dev/disk/by-path/pci-0000:00:1f.2-ata-1.0-part1 -> ../../sda1 lrwxrwxrwx 1 10 Mar 13 03:55 /dev/disk/by-uuid/3982f3a2-3e59-4658-b18d-5a036572c6c7 -> ../../sda1 So there's a hex 20 (space) in there, but it's happily escaped, so it may actually be the \x20 that is stored in the label name. Curious. Of course, I have no idea how it got there! I created a new gpt partition from scratch and then sliced it as shown diskfarm:~ # parted /dev/sda print Model: ATA SanDisk SD6SB1M1 (scsi) Disk /dev/sda: 128GB Sector size (logical/physical): 512B/512B Partition Table: gpt Disk Flags: pmbr_boot Number Start End Size File system Name Flags 1 1049kB 35.4GB 35.4GB linux-swap(v1) diskfarm-swap swap 2 35.4GB 69.8GB 34.4GB xfs diskfarmsuse 3 69.8GB 104GB 34.4GB diskfarmknop legacy_boot 4 104GB 128GB 23.9GB xfs diskfarm-ssd and I certainly didn't add "System Reserved" or "System\x20Reserved" anywhere. Sooooo ... wtf is it?!? Haven't yet had a chance to reboot after rebuilding diskfarm:~ # egrep -v '^$|^#' /etc/default/grub GRUB_DISTRIBUTOR= GRUB_DEFAULT=saved GRUB_HIDDEN_TIMEOUT=0 GRUB_HIDDEN_TIMEOUT_QUIET=true GRUB_TIMEOUT=2 # DTG 8 GRUB_CMDLINE_LINUX_DEFAULT="mitigations=auto quiet" GRUB_CMDLINE_LINUX="" GRUB_DISABLE_OS_PROBER='true' ### do not find the Win 7 restore image partition GRUB_TERMINAL="gfxterm" GRUB_GFXMODE="auto" GRUB_BACKGROUND= GRUB_THEME=/boot/grub2/themes/openSUSE/theme.txt SUSE_BTRFS_SNAPSHOT_BOOTING="true" GRUB_DISABLE_OS_PROBER="false" GRUB_ENABLE_CRYPTODISK="n" GRUB_CMDLINE_XEN_DEFAULT="vga=gfx-1024x768x16" my grub config. [BTW, setting GRUB_DISABLE_OS_PROBER had no effect *sigh* :-] Thanks again & HANN :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt
On 3/20/21 11:46 PM, David T-G wrote:
Actually, maybe not. When I look via ls, I see an encoded space:
That misses the point. There is a label "System Reserved" shown in the "resume=" boot parameter. But your error message (taken from first message in this thread) is: A start job is running for dev-disk-by\x2dlabel-System.device That does not include the "Reserved" part of the label. So the installer did something wrong. So a bug report is needed, so that the installer gets it right in future.
participants (7)
-
Carlos E. R. -
Carlos E.R. -
Dave Howorth -
David T-G -
Felix Miata -
J Leslie Turriff -
Neil Rickert