at command is not executing tasks

Reseko · 08-08-2023, 05:39 AM

Buenos días,

A few days ago, I installed Slackware 15 in a VM for practice (I'm learning), and I've noticed that the "at" command is not working.

I create any task, and it just stays there, it never gets executed. It doesn't matter which user I use to run it. I've tried with a non-privileged user and with root, and it's the same.

The "atd" daemon is running with the following parameters:

/usr/bin/atd -b 15 -l 1

Could someone lend me a hand or point me in the right direction?

Thank you.

Regards.

pan64 · 08-08-2023, 05:46 AM

you may need to check logs, or if you run it in a terminal you can use -d too.

metaed · 08-08-2023, 06:23 AM

Also check your load average.

Petri Kaukasoina · 08-08-2023, 06:38 AM

Quote:

Originally Posted by Reseko

I create any task, and it just stays there, it never gets executed.

How did you try it? For example:

Code:

echo 'touch /tmp/atworks' | at now + 2 minutes

atq shows the job in the queue. Does it not vanish after 2 minutes? Doesn't it create file /tmp/atworks ?

Quote:

Originally Posted by metaed

Also check your load average.

I think '-l 1' only limits jobs submitted using 'batch' to run with low enough load. 'at' jobs should start regardless.

GazL · 08-08-2023, 07:30 AM

at 3.2.4 (shipped in 15.0) was broken. Pat downgraded to at 3.2.3 in 'patches', so make sure you apply all the patches to your 15.0 system.

previous thread

Reseko · 08-09-2023, 04:02 AM

Quote:

Originally Posted by pan64

you may need to check logs, or if you run it in a terminal you can use -d too.

I have reviewed syslog and messages, and I have not seen any annotation related to atd.

Tasks simply remain in the at queue but do not execute.

Reseko · 08-09-2023, 04:03 AM

Quote:

Originally Posted by metaed

Also check your load average.

The system is freshly reinstalled and idle. Running a 'top' command shows minimal system load.

Reseko · 08-09-2023, 04:05 AM

Quote:

Originally Posted by Petri Kaukasoina

How did you try it? For example:

Code:

echo 'touch /tmp/atworks' | at now + 2 minutes

atq shows the job in the queue. Does it not vanish after 2 minutes? Doesn't it create file /tmp/atworks ?

Exactly, the task remains in the queue but does not disappear. It stays there indefinitely.

I think '-l 1' only limits jobs submitted using 'batch' to run with low enough load. 'at' jobs should start regardless.

The parameters are the ones set by default by Slackware, I didn't change anything.

Reseko · 08-09-2023, 04:49 AM

Quote:

Originally Posted by GazL

at 3.2.4 (shipped in 15.0) was broken. Pat downgraded to at 3.2.3 in 'patches', so make sure you apply all the patches to your 15.0 system.

previous thread

I updated using slackpkg, rebooted the system, and now the system is giving me the following error: "VMware Workstation unrecoverable error: (vcpu-0)."

I had another Slackware VM installed, repeated the same process, and on this one, I'm getting a different error

: "Loading LinuxEBDA is big; kernel setup stack overlaps LILO second stage."

I come from the Debian environment, and I also don't have much experience, so I don't know why all of this could be happening, haha . I'm used to using apt and not having any issues of any kind.

Greetings.

baumei · 08-09-2023, 07:32 AM

When you ran slackpkg, did you have it upgrade the kernel along with the other software?

jayjwa · 08-09-2023, 01:06 PM

It's broken here too now but used to work. Slackware current.

Code:

Aug  9 13:38:00 atr2 atd[2965]: pam_unix(atd:session): session opened for user jayjwa(uid=1000) by (uid=2)
Aug  9 13:38:00 atr2 atd[2965]: gkr-pam: unable to locate daemon control file
Aug  9 13:38:00 atr2 atd[2966]: gkr-pam: couldn't setup credentials: Operation not permitted
Aug  9 13:38:00 atr2 atd[2965]: gkr-pam: gnome-keyring-daemon didn't start properly
Aug  9 13:38:00 atr2 atd[2965]: pam_env(atd:setcred): Unrecognized Option: @{HOME}  - ignoring line

It might be because of PAM https://bugs.archlinux.org/task/61700 since there was no /etc/pam.d/atd. Even after adding one it still doesn't work.

Jobs just get stuck.

Code:

ls -l -a /var/spool/at*                                                                                                                                                                                         
/var/spool/atjobs:
total 16K
drwxrwx--T  2 daemon daemon 4.0K Aug  9 13:36 ./
drwxr-xr-x 18 root   root   4.0K Jul 19 15:21 ../
-rwx------  1 jayjwa daemon 2.9K Aug  9 13:36 a0000201ae3262*
-rw-------  1 daemon daemon    6 Aug  9 13:36 .SEQ

/var/spool/atspool:
total 8.0K
drwxrwx--T  2 daemon daemon 4.0K May 20 16:57 ./
drwxr-xr-x 18 root   root   4.0K Jul 19 15:21 ../

Code:

at now + 2 minutes
warning: commands will be executed using /bin/sh
at Wed Aug  9 13:38:00 2023
at> touch /tmp/atworks.txt
at> <EOT>
job 2 at Wed Aug  9 13:38:00 2023
jayjwa@atr2 ~> atq
2	Wed Aug  9 13:38:00 2023 a jayjwa

This is what I tried to add but it's not working.

Code:

cat /etc/pam.d/atd 
auth      required    pam_unix.so
auth      required    pam_env.so

account   required    pam_access.so
account   required    pam_unix.so
account   required    pam_time.so

session   required    pam_loginuid.so
session   required    pam_limits.so
session   required    pam_unix.so

henca · 08-09-2023, 01:11 PM

Simply just using slackpkg to update a system might seem convenient, but failing to read important instructions in the ChangeLog.txt will result in a non booting system:

Code:

+--------------------------+
Sat Jun 24 00:16:22 UTC 2023
patches/packages/linux-5.15.117/*:  Upgraded.
  We're going to back up one version to avoid an amdgpu regression in 5.15.118.
  If you're already using 5.15.118 without issues, feel free to stick with it.
  Be sure to upgrade your initrd after upgrading the kernel packages.
  If you use lilo to boot your machine, be sure lilo.conf points to the correct
  kernel and initrd and run lilo as root to update the bootloader.
  If you use elilo to boot your machine, you should run eliloconfig to copy the
  kernel and initrd to the EFI System Partition.
+--------------------------+

regards Henrik

Petri Kaukasoina · 08-09-2023, 01:20 PM

Quote:

Originally Posted by jayjwa

It's broken here too now but used to work. Slackware current.

Code:

Aug  9 13:38:00 atr2 atd[2965]: gkr-pam: unable to locate daemon control file
Aug  9 13:38:00 atr2 atd[2966]: gkr-pam: couldn't setup credentials: Operation not permitted
Aug  9 13:38:00 atr2 atd[2965]: gkr-pam: gnome-keyring-daemon didn't start properly

at works for me on current, but on the other hand, I have given 'removepkg gnome-keyring' and added gnome-keyring to slackpkg blacklist.

volkerdi · 08-09-2023, 02:11 PM

Working here on -current with all packages installed.

lostintime · 08-09-2023, 05:10 PM

Quote:

I have reviewed syslog and messages, and I have not seen any annotation related to atd.

Look in /var/log/secure for something like this:

Code:

atd[9626]: pam_unix(atd:session): session [opened|closed] for user

Quote:

I updated using slackpkg, rebooted the system

Verify version 3.23 is installed:

ls /var/lib/pkgtools/packages/at-3.2.*

The regression was the first 15.0 bug report resolved with a patched package:

Code:

Thu Feb 10 01:46:55 UTC 2022
patches/packages/at-3.2.3-x86_64-1_slack15.0.txz:  Upgraded.
  Switched to at-3.2.3 since version 3.2.4 has a regression that causes
  queued jobs to not always run on time when atd is run as a standalone
  daemon. Thanks to Cesare.

Quote:

...now the system is giving me the following error: "VMware Workstation unrecoverable error: (vcpu-0)."

Snooping around the web for that error message indicates at least a half dozen possible reasons.

Quote:

I had another Slackware VM installed, repeated the same process, and on this one, I'm getting a different error : "Loading LinuxEBDA is big; kernel setup stack overlaps LILO second stage."

Some wild guesses:

* Not running "lilo" after updating the kernel.
* Insufficient RAM allocated to the VM.
* Using a separate /boot partition and the partition is too small or full.