Fedora Linux

2147 readers

2 users here now

All about Fedora Linux

founded 5 years ago

MODERATORS

CAP_NAME_NOW_UPVOTE@lemmy.ml

[solved] smartd.service causing long boot times (lemmy.nz)

submitted 4 days ago* (last edited 4 days ago) by terraborra@lemmy.nz to c/fedora@lemmy.ml

8 comments fedilink hide all child comments

I'm running Fedora 41 KDE and had what appeared to be an issue with the system not loading to desktop from the GUI login screen. I am getting a black screen with just the cursor and nothing else. I can get a terminal window going on TTY3-6 to reboot and doing so would load into the desktop pretty much instantly.

Turns out that there's actually a process causing total boot time from power on to exceed 6 minutes. Boot log is viewable here https://paste.centos.org/view/raw/8846c1ec

Using systemd-analyze blame I can see that smartd.service is causing 3 minutes of that boot time. From a quick search it seems like it might be doing a full SMART test of my 3 SSDs on the first boot of the day.

Does anyone know how to disable this at boot? There doesn't appear to be a conf file in /etc/ and smartmontools isn't installed.

edit with solution

Thanks @just_another_person@lemmy.world and @notanapple@lemm.ee for correctly pointing out that fedora was struggling to mount a drive.

I checked fstab and there was no entry for the drive. mtab showed that it was mounted even though the drive wasn’t listed. I tried to find a way to have the drive ignored at boot but the only results that came up involved setting udev rules and those results were more than a decade old.

In the end I found the drive listed in /dev/. Using

rm nvmeXnXpX

I deleted the two partition entries and it now boots perfectly AND the drive is now listed correctly in the desktop and accessible.

top 8 comments

sorted by: hot top controversial new old

[–] just_another_person@lemmy.world 3 points 4 days ago* (last edited 4 days ago) (2 children)

From your log, this seems to be the issue:

kf.solid.backends.udisks2: Failed enumerating UDisks2 objects: "org.freedesktop.DBus.Error.NoReply" "Did not receive a reply. Possible causes include: the remote application did not send a reply, the message bus security policy blocked the reply, the reply timeout expired, or the network connection was broken."

Then it gets into a loop between smartd and udisks2. You sure you're drives are all okay? It really looks like the disk services are bailing because there is a problem. Did you remove a drive or something recently?

[–] terraborra@lemmy.nz 2 points 4 days ago* (last edited 4 days ago) (1 children)

Ah, I think that is a windows drive with games on it. I originally was using Pop OS and tried to boot games directly from that drive before learning that you had to do a fresh install of them in Linux.

After a certain point that drive no longer appeared on Pop and it doesn’t show in fedora at all. Possibly the path got corrupted.

I still have pop installed to the same drive that I’m booting fedora from. I’ll try wiping that part of the partition and see if it fixes it.

Thank you that helps me narrow it down as I was struggling to interpret the boot log.

[–] just_another_person@lemmy.world 3 points 4 days ago (1 children)

Well wiping the drive won't fix the issue entirely if there is some mapping for it existing. Check fstab, and any other mounts in KDE you may have for it and get rid of them. Should be super obvious on a reboot if you've fixed it by doing so.

[–] terraborra@lemmy.nz 3 points 4 days ago

Awesome thank you. Didn’t even know that fstab was a thing

[–] notanapple@lemm.ee 2 points 4 days ago (1 children)

could also be an issue with kf.solid.backends.udisks2

[–] just_another_person@lemmy.world 2 points 4 days ago* (last edited 4 days ago) (1 children)

Yes, one informs the other, hence the loop. It feels like there is a mount or something in fstab which has become invalid, and the service keeps getting stuck in a segfault>kill loop with no clean exit.

[–] notanapple@lemm.ee 2 points 4 days ago (1 children)

Saw this in the log:

Apr 04 13:47:27 kernel: nvme nvme1: Device not ready; aborting reset, CSTS=0x1
Apr 04 13:47:27 kernel: nvme nvme1: Disabling device after reset failure: -19
Apr 04 13:47:27 udisksd[2910]: Error probing device: NVMe Identify Namespace command error: Input/output error (g-bd-nvme-error-quark, 1)
Apr 04 13:47:27 kernel: nvme nvme1: Identify namespace failed (-5)
Apr 04 13:47:27 udisksd[2910]: Error performing housekeeping for drive /org/freedesktop/UDisks2/drives/INTEL_SSDPEKKW512G8_BTHH82250WSG512D: Error updating Health Information: No probed controller info available (udisks-error-quark, 0)

Seems you were right, there is some kind of device or connection issue.

[–] just_another_person@lemmy.world 2 points 4 days ago

Good eyes! I missed that completely.