Created attachment 857284 [details] Output from /proc/cmdline Description of problem: Running Fedora 20 as Guest OS in VirtualBox 4.3 (Host OS: OS X Mountain Lion), a stop-job-bug, causing a delay of approximately 1.5 minutes during the shutdown process, appears in a reproducible fashion when shutting down or rebooting using GUI controls or using certain commands, as elaborated below. In the last year, I have been running CentOS 6.3, CentOS 6.5, Fedora 18, Fedora 19, Debian 7.0 Wheezy, and Windows 7 Ultimate as Guest OSes in VirtualBox without encountering the stop-job-bug so the stop-job-bug is not a bug in VirtualBox. The stop-job-bug is not a kernel bug: I ran the commands `sync && reboot -f' and `sync && poweroff -f' as the root user several times, thereby bypassing the init system, and I didn't encounter any problems using those commands. I edited /etc/default/grub, removing the parameters 'rhgb' and 'quiet' from the value of GRUB_CMDLINE_LINUX so that the kernel messages were visible during boot, reboot, and shutdown. I ran `grub2-mkconfig -o /boot/grub2/grub.cfg' in order to update the grub2 configuration file (grub.cfg). Finally, I rebooted from the Xfce desktop using the command `init 6' and, luckily, the stop-job-bug appeared during the reboot. I captured a screen shot of the stop-job-bug screen using the Grab utility in OS X Mountain Lion. Please refer to stop_job_bug_reboot_995.tiff. The screen shot displayed the text: "A stop job is running for User Manager for 995". Afterwards, I booted Fedora 20 and ran the command `shutdown -h' from the Xfce desktop. The stop-job-bug reappeared during shutdown but the text now read: "A stop job is running for User Manager for 1000". Please refer to stop_job_bug_shutdown_1000.tiff. Every time I happened to reproduce the stop-job-bug during shutdown or reboot, the same two text messages appeared either referring to "User Manager for 995" or referring to "User Manager for 1000". Please refer to stop_job_bug_reboot_1000.tiff and stop_job_bug_shutdown_995.tiff in addition to the tiff files mentioned above. Then I ran the commands `grep '995' /var/log/messages | less' and `grep '1000' /var/log/messages | less' and the outputs from these commands suggest that the bug might be related to the new slice feature in systemd. Not every shutdown or reboot of Fedora 20 triggers the stop-job-bug. The stop-job-bug appears in a reproducible fashion when shutting down Fedora 20 using the GUI controls in Xfce or shutting down / rebooting using one of the following commands (run from the Xfce desktop (terminal emulator) or from one of the virtual consoles): # shutdown -h (from the Xfce desktop or from a virtual console) # shutdown -h now (from the Xfce desktop) # halt --poweroff (from the Xfce desktop or from a virtual console) # poweroff (from the Xfce desktop) # telinit 0 (from the Xfce desktop) # init 0 (from the Xfce desktop) # init 6 (from the Xfce desktop) # shutdown -r (from the Xfce desktop or from a virtual console) # systemctl poweroff (from the Xfce desktop or from a virtual console) # systemctl reboot (from the Xfce desktop) I have not been able to reproduce the stop-job-bug when shutting down or rebooting Fedora 20 using one of the following commands (run from the Xfce desktop (terminal emulator) or from one of the virtual consoles): # poweroff (from a virtual console) # shutdown (from the Xfce desktop or from a virtual console) # shutdown --poweroff (from the Xfce desktop or from a virtual console) # shutdown -h now (from a virtual console) # telinit 0 (from a virtual console) # init 0 (from a virtual console) # shutdown -r now (from the Xfce desktop or from a virtual console) # telinit 6 (from the Xfce desktop or from a virtual console) # init 6 (from a virtual console) # systemctl reboot (from a virtual console) # reboot (from the Xfce desktop or from a virtual console) A reboot using GUI controls from the login screen is unaffected by the stop-job-bug. But a reboot using GUI controls from the Xfce desktop triggers the stop-job-bug. When running `systemctl kexec', from a virtual console as well as from the Xfce desktop, kexec failed and I was dropped to a debug shell by dracut. For a screen shot of the warning from dracut and the debug shell, please refer to kexec.tiff. I investigated, however, the kexec issue no further as it seems to be quite another bug, probably a kernel bug. Attachments: (1) stop_job_bug_reboot_995.tiff (2) stop_job_bug_reboot_1000.tiff (3) stop_job_bug_shutdown_995.tiff (4) stop_job_bug_shutdown_1000.tiff (5) kexec.tiff (6) output from /proc/cmdline (file: cmdline) (7) output from the `dmesg' command (file: dmesg.txt), captured immediately after a reboot affected by the stop-job-bug; additional boot parameters: 'systemd.log_level=debug systemd.log_target=kmsg log_buf_len=1M' (8) a copy of /var/log/messages (file: messages) (9) output from the command `VBoxManage showvminfo "Fedora_20" --details' (file: virtualbox_config_info.txt) Version-Release number of selected component (if applicable): 208-9.fc20.i686 How reproducible: Reboot or shut down Fedora 20 using the GUI controls on the desktop, at the login screen or one of the commands: `shutdown', `halt', `poweroff', `telinit', `init', `reboot' or `systemctl'. Steps to Reproduce: 1. Boot Fedora 20 and log in to the desktop or a virtual console. 2. Reboot or shut down Fedora 20 using the GUI controls or one of the commands mentioned above. 3. Wait for the stop-job-bug to appear. NB: 'rhgb' and 'quiet' must be deleted from /etc/default/grub so that the kernel text messages appear during shutdown and reboot. Actual results: A delay of approximately 1.5 minutes during shutdown or reboot. Expected results: No delay during shutdown or reboot. Additional info: Please refer to the excruciatingly detailed problem description above.
Created attachment 857286 [details] Output from the `dmesg' command
Created attachment 857287 [details] A copy of /var/log/messages
Created attachment 857288 [details] VirtualBox configuration information
Created attachment 857290 [details] Warning from dracut and debug shell triggered by `systemctl kexec'
Created attachment 857291 [details] Stop-job-bug screen shot captured during reboot 1
Created attachment 857293 [details] Stop-job-bug screen shot captured during reboot 2
Created attachment 857294 [details] Stop-job-bug screen shot captured during shutdown 1
Created attachment 857295 [details] Stop-job-bug screen shot captured during shutdown 2
Created attachment 858309 [details] Stop-job-bug screen shot captured during shutdown 3
February 2, 2014: I just came across another variation of the stop-job-bug screen - still pertaining to the new slice feature in systemd (cfr. `grep -i 'slice' messages'). Please refer to 'Stop-job-bug screen shot captured during shutdown 3'.
Confirming. 4 VCPU Fedora 20 guest, running on RHEL-7.0 qemu-kvm. Seeing the message with "995".
Forgot to mention -- I'm also using the Xfce spin. Plus, # systemctl status user user - User Manager for 995 Loaded: loaded (/usr/lib/systemd/system/user@.service; static) Active: active (running) since Fri 2014-04-04 23:58:01 CEST; 1min 16s ago Main PID: 661 (systemd) Status: "Startup finished in 110ms." CGroup: /user.slice/user-995.slice/user |-661 /usr/lib/systemd/systemd --user `-664 (sd-pam) Apr 04 23:58:01 ovmf-f20 systemd[661]: pam_unix(systemd-user:session): session opened for user lightdm by (uid=0) Apr 04 23:58:01 ovmf-f20 systemd[1]: Started User Manager for 995. # id lightdm uid=995(lightdm) gid=993(lightdm) groups=993(lightdm) It could be a problem with LightDM.
Hi Troels, this should be fixed with systemd-208-14 or 208-15. In comment #1 you say you're running systemd-208-9. Can you try with the lastest one?
(In reply to Zbigniew Jędrzejewski-Szmek from comment #13) Hi Zbigniew: As of today, April 11 2014, I'm running systemd-208-15, and I'm no longer able to reproduce the bug that I reported. > Hi Troels, this should be fixed with systemd-208-14 or 208-15. In comment #1 > you say you're running systemd-208-9. Can you try with the lastest one?
I can see something similar on F20 (fully up to date), but my machine is a laptop, not a virtual guest. Once in a while it happens and then shutdown takes ages
This message is a reminder that Fedora 20 is nearing its end of life. Approximately 4 (four) weeks from now Fedora will stop maintaining and issuing updates for Fedora 20. It is Fedora's policy to close all bug reports from releases that are no longer maintained. At that time this bug will be closed as EOL if it remains open with a Fedora 'version' of '20'. Package Maintainer: If you wish for this bug to remain open because you plan to fix it in a currently maintained version, simply change the 'version' to a later Fedora version. Thank you for reporting this issue and we are sorry that we were not able to fix it before Fedora 20 is end of life. If you would still like to see this bug fixed and are able to reproduce it against a later version of Fedora, you are encouraged change the 'version' to a later Fedora version prior this bug is closed as described in the policy above. Although we aim to fix as many bugs as possible during every release's lifetime, sometimes those efforts are overtaken by events. Often a more recent Fedora release includes newer upstream software that fixes bugs or makes them obsolete.
Fedora 20 changed to end-of-life (EOL) status on 2015-06-23. Fedora 20 is no longer maintained, which means that it will not receive any further security or bug fix updates. As a result we are closing this bug. If you can reproduce this bug against a currently maintained version of Fedora please feel free to reopen this bug against that version. If you are unable to reopen this bug, please file a new report against the current release. If you experience problems, please add a comment to this bug. Thank you for reporting this bug and we are sorry it could not be fixed.