Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
Cant mount partitions from fstab timed out on fresh inst
View unanswered posts
View posts from last 24 hours

 
Reply to topic    Gentoo Forums Forum Index Installing Gentoo
View previous topic :: View next topic  
Author Message
Pipeng
Tux's lil' helper
Tux's lil' helper


Joined: 23 Jul 2013
Posts: 87

PostPosted: Fri Nov 24, 2023 10:52 pm    Post subject: Cant mount partitions from fstab timed out on fresh inst Reply with quote

Hi all, I have installed Gentoo in a machine, but I have some very weird things happening.

The installation want pretty well, even almost all works! except for some final details, one of them is:

I have installed on a MZ32-AR0 V.1.0 mobo a NVME drive, which is has the next partitions, Boot, EFI, Backup SO, Gentoo and Swap.

In order to mount it I tested a fstab configuration I had on other place that also has NVME, how didn't works I tested use genfstab, which give me the same result.

The problem is the next one:

Code:

nov 24 19:04:58 localhost systemd[1]: dev-disk-by\x2ddiskseq-1\x2dpart5.device: Job dev-disk-by\x2ddiskseq-1\x2dpart5.device/start timed out.
nov 24 19:04:58 localhost systemd[1]: Timed out waiting for device dev-disk-by\x2ddiskseq-1\x2dpart5.device.
nov 24 19:04:58 localhost systemd[1]: Dependency failed for dev-disk-by\x2ddiskseq-1\x2dpart5.swap.
nov 24 19:04:58 localhost systemd[1]: dev-disk-by\x2ddiskseq-1\x2dpart5.swap: Job dev-disk-by\x2ddiskseq-1\x2dpart5.swap/start failed with result 'dependency'.
nov 24 19:04:58 localhost systemd[1]: dev-disk-by\x2ddiskseq-1\x2dpart5.device: Job dev-disk-by\x2ddiskseq-1\x2dpart5.device/start failed with result 'timeout'.


Here systemd is trying to mount automatically the swap partition, which I have disabled.

The fstab:
Code:

# /dev/nvme0n1p4
UUID=573869ca-80a2-43dc-ba05-7c554558b89d   /            ext4         rw,noatime,x-systemd.device-timeout=600   0 1

# portal
#portal                 /run/user/1000/doc   fuse.portal   rw,nosuid,nodev,user_id=1000,group_id=1000   0 0

# tracefs
#tracefs                /sys/kernel/tracing   tracefs      rw,nosuid,nodev,noexec   0 0

# /dev/nvme0n1p1
#UUID=d60abf33-5ba8-4689-bfb9-4bc87a023f6e   /boot        ext4         rw,relatime,x-systemd.device-timeout=600   0 2

# /dev/nvme0n1p2
#UUID=8DA2-C0AD         /boot/efi    vfat         rw,relatime,fmask=0022,dmask=0022,codepage=437,iocharset=ascii,shortname=mixed,errors=remount-ro,x-systemd.device-timeout=600   0 2

# /dev/nvme0n1p5
#UUID=0c0b56b7-9040-4ec3-88c5-76771ffe82b2   none         swap         defaults     0 0


Almost all disks disabled because fails with the same error as above, the only one that works is /.

Some answers:

Have you tested increase mount time?

Yes with 5mins of max, same result.

Seems you are using NVME, maybe you should set CONFIG_BLK_DEV_NVME=y

I tried, after that the kernel was unable to find any drive or partition, need to rollback that.

Can you mount the partitions manually?

Yes, even if systemd fails I can manually mount /boot, /boot/efi and swap without problem.

Have you checked if the partitions are right?

Almost all of them, except root all of them has been checked, also, how I'm running from / and there is no errors about that on journalctl should be fine.

What is your kernel config?

https://gist.github.com/latot/220452d77efa62dcb86e50f5d654604a

Reading the error, seems systemd is trying to read using diskseq, can you show that folder?

Here it is:
Code:

localhost /dev/disk # dir
by-id  by-partuuid   by-uuid


And, if you wait some time, by reasons I don't know it appears.

Code:

localhost /dev/disk # dir
by-diskseq  by-id  by-partuuid   by-uuid


Maybe does not exists on boot time and that is why the timeout.

Have you tried to rebuild the kernel?

Yes, a lot of time, several kernels tested, same result.

You need to run systemctl daemon-reload to apply changes on fstab

Yes, I have been doing that.

Which kernel and systemd are you running?

systemd-254.5-r1
gentoo-sources-6.1.57-gentoo-x86_64


Now no more ideas in how to do this... someone has idea why this happens?
Back to top
View user's profile Send private message
alamahant
Advocate
Advocate


Joined: 23 Mar 2019
Posts: 3879

PostPosted: Sat Nov 25, 2023 10:25 am    Post subject: Reply with quote

Try removing
Code:

x-systemd.device-timeout=600

from fstab.
Also try with
gentoo-kernel-bin
just to be sure its not kernel related.
Can you boot into your system despite the errors?
Can you plz post
Code:

grep -i cmd /etc/default/grub

_________________
:)
Back to top
View user's profile Send private message
Pipeng
Tux's lil' helper
Tux's lil' helper


Joined: 23 Jul 2013
Posts: 87

PostPosted: Sat Nov 25, 2023 8:39 pm    Post subject: Reply with quote

Hi! the timeout I added it to just test, sorry I forgot to remove it, is the same if is there or not.

The grub config:
Code:

#GRUB_CMDLINE_LINUX=""
# GRUB_CMDLINE_LINUX="net.ifnames=0"
# GRUB_CMDLINE_LINUX="init=/usr/lib/systemd/systemd"
#GRUB_CMDLINE_LINUX_DEFAULT=""


About the kernel, new clues! test with the bin one was a great idea.

I'll call kernel-bin the one you wrote, it works!, but no so simple, the stable one is on 6.1.60 and sources are in 6.1.57, so I perform the next tests using the kernel-bin config and mixing versions.

sources 6.1.57 + my custom config = FAIL
sources 6.1.57 + kernel-bin_config = FAILS

kernel-bin 6.1.60 + kernel-bin_config = WORKS
sources 6.1.60 + kernel_bin-config = WORKS
sources 6.1.60 + my custom config = FAIL

So, how 6.1.57 fails with both config, is more like there is a bug/issue between 57-60 that could be fixed or a new options.

How in 6.1.60 works with both kernels using the default config, means there is no problem with gentoo-sources.

Now... the kernel config I build fails, I followed the docs so there should not be major changes, so how works with the kernel_bin-config would means there can be a new option in kernel, that fixes this, now... from there to know which option is would be quiet a chalange..

Any ideas how to proceed?

Sources of kernel-bin:
https://gist.github.com/latot/8241d1ec56ed1294aafa4e2212115242
Back to top
View user's profile Send private message
CooSee
Veteran
Veteran


Joined: 20 Nov 2004
Posts: 1444
Location: Earth

PostPosted: Sat Nov 25, 2023 9:57 pm    Post subject: Reply with quote

IIRC filesystem specific devices must be set to YES, e.g:

Code:
#
# NVME Support
#
CONFIG_NVME_CORE=y
CONFIG_BLK_DEV_NVME=y
# CONFIG_NVME_MULTIPATH is not set
# CONFIG_NVME_VERBOSE_ERRORS is not set
# CONFIG_NVME_FC is not set
# CONFIG_NVME_TCP is not set
# CONFIG_NVME_AUTH is not set
# CONFIG_NVME_TARGET is not set
# end of NVME Support

also, the following to YES.
Code:
CONFIG_SATA_AHCI=y

the above is from my .config - you've set all as Modules.

8)
_________________
" Die Realität ist eine Illusion, die durch Mangel an ehrlicher Kommunikation entsteht "
---
" Der Mensch ist von Natur aus neugierig, was am Ende übrig bleibt ist die Gier "
Back to top
View user's profile Send private message
Pipeng
Tux's lil' helper
Tux's lil' helper


Joined: 23 Jul 2013
Posts: 87

PostPosted: Sun Nov 26, 2023 3:17 am    Post subject: Reply with quote

Hi!, as described above use

Code:
CONFIG_BLK_DEV_NVME=y

causes to do not detect any device, O thought with
Code:
CONFIG_SATA_AHCI=y

would works, sadly was the same result.

But, I checked the fails having it enabled, helped to found this post:
https://forums.gentoo.org/viewtopic-t-813229-start-0.html

The option
Code:
CONFIG_SYSFS_DEPRECATED

if is enabled will block the creation/mount of partitions, dues to that no disk will be detected with
Code:
CONFIG_BLK_DEV_NVME=y
.

Also seems due to this the disks does not exists for systemd to mount them with fstab.

Code:
CONFIG_BLK_DEV_NVME=m

Can be as module.

But all to works
Code:
CONFIG_SYSFS_DEPRECATED=n

must be disabled.

I was going to report in other issue gnome-disks does not shows any disk while blkid do, doing this change also fixed this.

Why was the option enable? I enabled that option following the Gentoo docs, there should be one very old part that are not updated.

Thx
Back to top
View user's profile Send private message
Hu
Administrator
Administrator


Joined: 06 Mar 2007
Posts: 21708

PostPosted: Sun Nov 26, 2023 4:17 pm    Post subject: Reply with quote

Your original kernel configuration has SYSFS_DEPRECATED=y. Your solution post says that SYSFS_DEPRECATED=n must be disabled for this to work. To me, this means that you need SYSFS_DEPRECATED=y, but that is what you had when it did not work. Did you mean that SYSFS_DEPRECATED=n is required?

Note that SYSFS_DEPRECATED is removed in newer kernels, as of driver core: remove CONFIG_SYSFS_DEPRECATED and CONFIG_SYSFS_DEPRECATED_V2, so updating to a newer kernel would also have produced the effect of setting SYSFS_DEPRECATED=n.

As I understand the removal message, no one on Gentoo should have set SYSFS_DEPRECATED=y for many years now. Any documentation instructing you to do so is wrong and needs to be changed.
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Installing Gentoo All times are GMT
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum