a start job is running for dev-disk-by\x2uuid...

[Laurentius]

There it is

Code: Select all

canelon3@debtest:~$ dmesg | grep -A8 ', max U'
[    7.573888] ata1.00: ATA-8: SAMSUNG HM160HI, HH100-15, max UDMA7
[    7.573891] ata1.00: 312581808 sectors, multi 0: LBA48 NCQ (depth 31/32), AA
[    7.576913] ata1.00: ACPI cmd 00/00:00:00:00:00:a0 (NOP) rejected by device (Stat=0x51 Err=0x04)
[    7.577104] ata1.00: ACPI cmd ef/10:01:00:00:00:a0 (SET FEATURES) rejected by device (Stat=0x51 Err=0x04)
[    7.577108] ata1.00: ACPI cmd ef/10:03:00:00:00:a0 (SET FEATURES) filtered out
[    7.577298] ata1.00: ACPI cmd ef/10:04:00:00:00:a0 (SET FEATURES) rejected by device (Stat=0x51 Err=0x04)
[    7.577490] ata1.00: ACPI cmd ef/10:05:00:00:00:a0 (SET FEATURES) rejected by device (Stat=0x51 Err=0x04)
[    7.581136] ata1.00: configured for UDMA/133
[    7.581267] scsi 0:0:0:0: Direct-Access     ATA      SAMSUNG HM160HI  0-15 PQ: 0 ANSI: 5
--
[    8.431413] ata5.00: ATAPI: TSSTcorp CDDVDW TS-L633M, 0200, max UDMA/100
[    8.450902] ata5.00: ACPI cmd 00/00:00:00:00:00:a0 (NOP) rejected by device (Stat=0x51 Err=0x04)
[    8.451191] ata5.00: ACPI cmd ef/10:01:00:00:00:a0 (SET FEATURES) succeeded
[    8.451470] ata5.00: ACPI cmd ef/10:02:00:00:00:a0 (SET FEATURES) succeeded
[    8.451473] ata5.00: ACPI cmd ef/10:03:00:00:00:a0 (SET FEATURES) filtered out
[    8.451751] ata5.00: ACPI cmd ef/10:04:00:00:00:a0 (SET FEATURES) succeeded
[    8.469430] ata5.00: configured for UDMA/100
[    8.472023] scsi 4:0:0:0: CD-ROM            TSSTcorp CDDVDW TS-L633M  0200 PQ: 0 ANSI: 5
[    8.800023] ata6: SATA link down (SStatus 0 SControl 300)
canelon3@debtest:~$ 

ata1.00 seems to be the hard drive. It doesen't make any clunks and seem to work well.
ata5.00 is the dvd reader/writer. it works very bad. barely can read anything and clunks a lot. i never use it and use an external usb dvd/rw

should i replace the hard drive?

may be related to this that sometimes, not very often, the system hangs for a while, may be a minute, may be three.... The GUI gets frozen for some time.

[bw123]

should i replace the hard drive?

The way I've narrowed down things like this is disconnect all drives, install a known good (or new) drive with os on a new cable, and zero fill and use the suspect drive as storage for a while. Run smartctl daily and use diff on today and yesterday's report.

may be related to this that sometimes, not very often, the system hangs for a while, may be a minute, may be three.... The GUI gets frozen for some time.

Yeah, that sounds typical for a failing drive. You might be able to straighten it out, but drives are so cheap now, you could get a brand new 500 gig for an hours wages. It would probably take you at least an hour to backup, zero fill, and restore the thing. If it's going to die an a month anyway, or keep acting like this, why spend the time?

[debiman]

ata5.00 is the dvd reader/writer. it works very bad. barely can read anything and clunks a lot. i never use it and use an external usb dvd/rw

i would DEFINITELY remove/disconnect the dvd.
at the very least it will help you narrow down your problem troubleshooting.

[Deb-fan]

This one is beyond me, it's all over the place issues. Can say every time I've encountered this systemd start job pita, it's been due to problems in fstab. So triple check your fstab uuids against the output of "sudo blkid".

Da mighty Hoas(Head-on-a-etc) recently advised/informed me of systemd's ability to automount swap even w/o it having an entry in fstab. Yep works, removed swap from fstab, still mounts fine.

Why yours doesn't is odd. Also just recently ran up against an old headache that may be relevant. When you install a new distro, some of their installers will format the swap partition, thus changing its uuids.

This just happened to me and the result was systemd running a 90sec start job every boot. Am not a systemd guru by any means but learned enough now, that whenever this start job junk happens, checking fstab is 1st stop.

More fstab weirdness, the no new new line thing. Not sure if this ever caused a systemd start job but has caused boot probs. What's a new new line you may ask ? You go to the end of your last entry in fstab(with the cursor) and hit Return key to jump down to the next empty line in the file. You just made a new new line in fstab. Doubt this is your issue just being anal.

Might try deleting your present swap w gparted and create a new one in its place. Though yeah that means you'll have to correct fstab for it in all the on board os's.

Hope you get it sorted, an ailing gnu/nix install can cause someone much distress.

PS, another random thought, might upgrade systemd from backports presently v230-7etc. May not do a thing for your issue but for something as important as init I wanted newer. Working under the assumption that newer versions will have squashed bugs and added or polished features.

[sunrat]

...Also just recently ran up against an old headache that may be relevant. When you install a new distro, some of their installers will format the swap partition, thus changing its uuids. ...

I had the same issue once. Easy fix once I worked out what the issue was. Edit fstab to reflect the new uuid as shown in blkid.
journalctl was useful to detect the issue.

[acewiza]

Please excuse my feeble attempt to change the topic, but Is anybody yet prepared to declare the "start job running" behavior a friendly, welcome feature of systemd?

[GarryRicketson]

Not me, but I can't speak for others,...
I can say though, I have encountered the same/similar error, fortunately not
very often,...but in any event it has and can occur on non systemd systems
as well, Old Stable , Debian 7 wheezy,...I even have seen it on Debian 6

by sunrat » I had the same issue once. Easy fix once I worked out what the issue was. Edit fstab to reflect the new uuid as shown in blkid.
journalctl was useful to detect the issue.

The same worked for me, but I did not have or use 'journalctl ' to detect it.

Code: Select all

~$ journalctl
bash: journalctl: command not found
garry@debian:~$ man
What manual page do you want?
garry@debian:~$ man journalctl
No manual entry for journalctl
 

[/size]
But editing fstab worked.

by acewiza » Please excuse my feeble attempt to change the topic

Nice try though

[acewiza]

I did not ever see that particular error message until I was up to Jessie so, my WAG was a systemd behavior of some sort - obviously due to a startup configuration (likely hardware) the system was having a problem with...

[Head_on_a_Stick]

Is anybody yet prepared to declare the "start job running" behavior a friendly, welcome feature of systemd?

Yes

EDIT: also, rather than editing /etc/fstab every time the swap partition is reformatted, the swap line can instead be removed for GPT drives and systemd will then automount the partition irrespective of the UUID.

[Deb-fan]

^Which is kinda nice and thanks for informing me of this systemd feature Hoas.