[opensuse] Talking of failing hibernation
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, The other day, talking about hibernation, I said that sometimes hibernation would simply fail for me. Well, this night my computer failed to hibernate - this is the log: <22>1 2017-12-18T04:28:37.783421+01:00 Telcontar clamd 3884 - - SelfCheck: Database status OK. <30>1 2017-12-18T04:29:23.150889+01:00 Telcontar systemd 1 - - Reached target Sleep. <30>1 2017-12-18T04:29:23.150889+01:00 Telcontar systemd 1 - - Reached target Sleep. <30>1 2017-12-18T04:29:23.172199+01:00 Telcontar systemd 1 - - Starting Hibernate... <7>1 2017-12-18T04:29:23.179861+01:00 Telcontar kernel - - - [569194.797529] PM: Hibernation mode set to 'platform' <30>1 2017-12-18T04:29:23.191726+01:00 Telcontar systemd-sleep 16509 - - INFO: running /usr/lib/systemd/system-sleep/grub2.sleep for hibernate <30>1 2017-12-18T04:29:23.191973+01:00 Telcontar systemd-sleep 16509 - - INFO: Running prepare-grub .. <30>1 2017-12-18T04:29:23.192395+01:00 Telcontar systemd-sleep 16509 - - 2017-12-18 04:29:23+01:00 - Hibernating the system now... <30>1 2017-12-18T04:29:23.198353+01:00 Telcontar systemd 1 - - Stopping Deferred execution scheduler... <28>1 2017-12-18T04:29:23.202597+01:00 Telcontar systemd-sh - - - Hibernating the system now... <30>1 2017-12-18T04:29:23.219928+01:00 Telcontar systemd-sleep 16509 - - service: no such service upsd.service <30>1 2017-12-18T04:29:23.220710+01:00 Telcontar systemd 1 - - Stopped Deferred execution scheduler. <30>1 2017-12-18T04:29:23.512897+01:00 Telcontar systemd-sleep 16509 - - running kernel is grub menu entry openSUSE Leap 42.2 (vmlinuz-4.4.92-18.36-default) <30>1 2017-12-18T04:29:23.513124+01:00 Telcontar systemd-sleep 16509 - - preparing boot-loader: selecting entry openSUSE Leap 42.2, kernel /boot/4.4.92-18.36-default <30>1 2017-12-18T04:29:29.227808+01:00 Telcontar systemd-sleep 16509 - - running /usr/sbin/grub2-once "openSUSE Leap 42.2" <30>1 2017-12-18T04:29:29.371469+01:00 Telcontar systemd-sleep 16509 - - time needed for sync: 5.7 seconds, time needed for grub: 0.1 seconds. <30>1 2017-12-18T04:29:29.371708+01:00 Telcontar systemd-sleep 16509 - - INFO: Done. <30>1 2017-12-18T04:29:29.371854+01:00 Telcontar systemd-sleep 16509 - - Suspending system... <6>1 2017-12-18T04:29:41.548570+01:00 Telcontar kernel - - - [569202.198976] PM: Syncing filesystems ... done. <6>1 2017-12-18T04:29:41.747276+01:00 Telcontar kernel - - - [569202.767822] Freezing user space processes ... (elapsed 0.016 seconds) done. <7>1 2017-12-18T04:29:41.747281+01:00 Telcontar kernel - - - [569202.784257] PM: Marking nosave pages: [mem 0x00000000-0x00000fff] <7>1 2017-12-18T04:29:41.747282+01:00 Telcontar kernel - - - [569202.784259] PM: Marking nosave pages: [mem 0x0009f000-0x000fffff] <7>1 2017-12-18T04:29:41.747283+01:00 Telcontar kernel - - - [569202.784262] PM: Marking nosave pages: [mem 0xbff90000-0xffffffff] <7>1 2017-12-18T04:29:41.747284+01:00 Telcontar kernel - - - [569202.784866] PM: Basic memory bitmaps created <6>1 2017-12-18T04:29:41.747286+01:00 Telcontar kernel - - - [569202.784888] PM: Preallocating image memory... <7>1 2017-12-18T04:29:41.747287+01:00 Telcontar kernel - - - [569212.924919] PM: Basic memory bitmaps freed <6>1 2017-12-18T04:29:41.747288+01:00 Telcontar kernel - - - [569212.924928] Restarting tasks ... done. <28>1 2017-12-18T04:29:41.820886+01:00 Telcontar rtkit-daemon 4543 - - The canary thread is apparently starving. Taking action. <29>1 2017-12-18T04:29:41.940237+01:00 Telcontar systemd 1 - - systemd-hibernate.service: Main process exited, code=exited, status=1/FAILURE <30>1 2017-12-18T04:29:41.941655+01:00 Telcontar rtkit-daemon 4543 - - Demoting known real-time threads. <27>1 2017-12-18T04:29:41.941831+01:00 Telcontar systemd 1 - - Failed to start Hibernate. <29>1 2017-12-18T04:29:41.950777+01:00 Telcontar rtkit-daemon 4543 - - Successfully demoted thread 4549 of process 4542 (/usr/bin/pulseaudio). <28>1 2017-12-18T04:29:41.950978+01:00 Telcontar systemd 1 - - Dependency failed for Hibernate. <29>1 2017-12-18T04:29:41.951114+01:00 Telcontar rtkit-daemon 4543 - - Successfully demoted thread 4548 of process 4542 (/usr/bin/pulseaudio). <29>1 2017-12-18T04:29:41.964488+01:00 Telcontar rtkit-daemon 4543 - - Successfully demoted thread 4542 of process 4542 (/usr/bin/pulseaudio). <29>1 2017-12-18T04:29:41.964740+01:00 Telcontar rtkit-daemon 4543 - - Demoted 3 threads. <28>1 2017-12-18T04:29:41.964870+01:00 Telcontar console-kit-daemon 4445 - - WARNING: Error waiting for native console 12 activation: Resource temporarily unavailable <28>1 2017-12-18T04:29:41.970606+01:00 Telcontar console-kit-daemon 4445 - - WARNING: Error waiting for native console 5 activation: Resource temporarily unavailable <30>1 2017-12-18T04:29:42.037062+01:00 Telcontar console-kit-daemon 4445 - - console-kit-daemon[4445]: WARNING: Error waiting for native console 12 activation: Resource temporarily unavailable <30>1 2017-12-18T04:29:42.037261+01:00 Telcontar console-kit-daemon 4445 - - console-kit-daemon[4445]: WARNING: Error waiting for native console 5 activation: Resource temporarily unavailable <29>1 2017-12-18T04:29:43.069508+01:00 Telcontar systemd 1 - - hibernate.target: Job hibernate.target/start failed with result 'dependency'. <30>1 2017-12-18T04:29:43.069771+01:00 Telcontar systemd 1 - - sleep.target: Unit not needed anymore. Stopping. <38>1 2017-12-18T04:29:43.078001+01:00 Telcontar systemd-logind 1706 - - Operation 'sleep' finished. <29>1 2017-12-18T04:29:43.081947+01:00 Telcontar systemd 1 - - systemd-hibernate.service: Unit entered failed state. <28>1 2017-12-18T04:29:43.082123+01:00 Telcontar systemd 1 - - systemd-hibernate.service: Failed with result 'exit-code'. <30>1 2017-12-18T04:29:43.091305+01:00 Telcontar systemd 1 - - Stopped target Sleep. <86>1 2017-12-18T04:30:01.946295+01:00 Telcontar cron 16841 - - pam_unix(crond:session): session opened for user cer by (uid=0) <86>1 2017-12-18T04:30:01.947130+01:00 Telcontar cron 16840 - - pam_unix(crond:session): session opened for user wwwrun by (uid=0) <86>1 2017-12-18T04:30:01.947316+01:00 Telcontar cron 16839 - - pam_unix(crond:session): session opened for user root by (uid=0) The reason is not clear: systemd 1 - - Dependency failed for Hibernate. systemd-hibernate.service: Failed with result 'exit-code'. systemd-hibernate.service: Main process exited, code=exited, status=1/FAILURE Well, I know that it failed, but why? Telcontar:~ # systemctl status systemd-hibernate ● systemd-hibernate.service - Hibernate Loaded: loaded (/usr/lib/systemd/system/systemd-hibernate.service; static; vendor preset: disabled) Active: failed (Result: exit-code) since Mon 2017-12-18 04:29:41 CET; 7h ago Docs: man:systemd-suspend.service(8) Process: 16509 ExecStart=/usr/lib/systemd/systemd-sleep hibernate (code=exited, status=1/FAILURE) Main PID: 16509 (code=exited, status=1/FAILURE) Dec 18 04:29:23 Telcontar systemd-sleep[16509]: running kernel is grub menu entry openSUSE Leap 42.2 (vmlinuz-4.4.92-18.36-default) Dec 18 04:29:23 Telcontar systemd-sleep[16509]: preparing boot-loader: selecting entry openSUSE Leap 42.2, kernel /boot/4.4.92-18.36-default Dec 18 04:29:29 Telcontar systemd-sleep[16509]: running /usr/sbin/grub2-once "openSUSE Leap 42.2" Dec 18 04:29:29 Telcontar systemd-sleep[16509]: time needed for sync: 5.7 seconds, time needed for grub: 0.1 seconds. Dec 18 04:29:29 Telcontar systemd-sleep[16509]: INFO: Done. Dec 18 04:29:29 Telcontar systemd-sleep[16509]: Suspending system... Dec 18 04:29:41 Telcontar systemd[1]: systemd-hibernate.service: Main process exited, code=exited, status=1/FAILURE Dec 18 04:29:41 Telcontar systemd[1]: Failed to start Hibernate. Dec 18 04:29:43 Telcontar systemd[1]: systemd-hibernate.service: Unit entered failed state. Dec 18 04:29:43 Telcontar systemd[1]: systemd-hibernate.service: Failed with result 'exit-code'. Telcontar:~ # I know that if I try again it will very probably succeed, but I would like to know why it failed this time. And of course, on the terminal that I called the command to hibernate, I got no error printed, it just returns as usual. This is the command I use: Telcontar:~ # cat /usr/local/bin/hibernate #!/bin/bash date --rfc-3339=seconds systemctl hibernate sleep 3 xscreensaver-command -lock Telcontar:~ # And this was the result: Telcontar:~ # hibernate 2017-12-18 04:29:23+01:00 xscreensaver-command: activating and locking. Telcontar:~ # - -- Cheers Carlos E. R. (from 42.2 x86_64 "Malachite" at Telcontar) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iEYEARECAAYFAlo3obwACgkQtTMYHG2NR9UAnQCdHt/LyoJF6h+kOSDK45BYXM3D Q58AoI76syr5PxRMbeIiVF1AIK5qFlhS =UmHM -----END PGP SIGNATURE-----
Le 18/12/2017 à 12:08, Carlos E. R. a écrit :
<28>1 2017-12-18T04:29:41.820886+01:00 Telcontar rtkit-daemon 4543 - - The canary thread is apparently starving. Taking action. <29>1 2017-12-18T04:29:41.940237+01:00 Telcontar systemd 1 - - systemd-hibernate.service: Main process exited, code=exited, status=1/FAILURE
what is this "canary" thread? jdd -- http://dodin.org -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
jdd@dodin.org wrote:
Le 18/12/2017 à 12:08, Carlos E. R. a écrit :
<28>1 2017-12-18T04:29:41.820886+01:00 Telcontar rtkit-daemon 4543 - - The canary thread is apparently starving. Taking action. <29>1 2017-12-18T04:29:41.940237+01:00 Telcontar systemd 1 - - systemd-hibernate.service: Main process exited, code=exited, status=1/FAILURE
what is this "canary" thread?
jdd
https://askubuntu.com/questions/86974/what-is-the-canary-thread-and-why-is-i... -- Per Jessen, Zürich (0.4°C) http://www.hostsuisse.com/ - virtual servers, made in Switzerland. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Monday, 2017-12-18 at 12:52 +0100, Per Jessen wrote:
jdd@dodin.org wrote:
Le 18/12/2017 à 12:08, Carlos E. R. a écrit :
<28>1 2017-12-18T04:29:41.820886+01:00 Telcontar rtkit-daemon 4543 - - The canary thread is apparently starving. Taking action. <29>1 2017-12-18T04:29:41.940237+01:00 Telcontar systemd 1 - - systemd-hibernate.service: Main process exited, code=exited, status=1/FAILURE
what is this "canary" thread?
jdd
https://askubuntu.com/questions/86974/what-is-the-canary-thread-and-why-is-i...
Interesting: |> So apparently some thread(s) that got real-time priorities from rtkit |> is/are misbehaving, and trying to monopolize the CPU, rtkit detects |> this with its "canary thread", and thus rtkit takes away the real-time |> priorities. At some point during hibernation processes are stopped, frozen. So the "canary" is also stopped, and apparently rtkit-daemon notices and "takes action" (mistakenly). This is, I hope, irrelevant because the message happens always: Telcontar:~ # grep "The canary thread is apparently starving." /var/log/messages <3.4> 2017-12-06 11:23:38 Telcontar rtkit-daemon 4543 - - The canary thread is apparently starving. Taking action. <3.4> 2017-12-07 11:23:11 Telcontar rtkit-daemon 4543 - - The canary thread is apparently starving. Taking action. <3.4> 2017-12-08 12:59:10 Telcontar rtkit-daemon 4543 - - The canary thread is apparently starving. Taking action. <3.4> 2017-12-08 22:06:59 Telcontar rtkit-daemon 4543 - - The canary thread is apparently starving. Taking action. <3.4> 2017-12-09 09:08:45 Telcontar rtkit-daemon 4543 - - The canary thread is apparently starving. Taking action. <3.4> 2017-12-09 21:06:16 Telcontar rtkit-daemon 4543 - - The canary thread is apparently starving. Taking action. <3.4> 2017-12-10 20:12:22 Telcontar rtkit-daemon 4543 - - The canary thread is apparently starving. Taking action. <3.4> 2017-12-11 21:41:29 Telcontar rtkit-daemon 4543 - - The canary thread is apparently starving. Taking action. <3.4> 2017-12-13 13:17:27 Telcontar rtkit-daemon 4543 - - The canary thread is apparently starving. Taking action. <3.4> 2017-12-14 13:05:27 Telcontar rtkit-daemon 4543 - - The canary thread is apparently starving. Taking action. <3.4> 2017-12-15 22:56:03 Telcontar rtkit-daemon 4543 - - The canary thread is apparently starving. Taking action. <3.4> 2017-12-16 11:53:45 Telcontar rtkit-daemon 4543 - - The canary thread is apparently starving. Taking action. <3.4> 2017-12-17 11:29:07 Telcontar rtkit-daemon 4543 - - The canary thread is apparently starving. Taking action. <3.4> 2017-12-18 04:29:41 Telcontar rtkit-daemon 4543 - - The canary thread is apparently starving. Taking action. Telcontar:~ # - -- Cheers, Carlos E. R. (from openSUSE 42.2 x86_64 "Malachite" at Telcontar) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iEYEARECAAYFAlo3snoACgkQtTMYHG2NR9XfjACffNhZDtetsTFWK9YVd0ew9AZR aLQAnRscOib7SMt0YHNRkQ4PB0V8o9nY =WAVj -----END PGP SIGNATURE-----
Carlos E. R. wrote:
Hi,
The other day, talking about hibernation, I said that sometimes hibernation would simply fail for me. Well, this night my computer failed to hibernate - this is the log:
<22>1 2017-12-18T04:28:37.783421+01:00 Telcontar clamd 3884 - - SelfCheck: Database status OK. <30>1 2017-12-18T04:29:23.150889+01:00 Telcontar systemd 1 - - Reached target Sleep. <30>1 2017-12-18T04:29:23.150889+01:00 Telcontar systemd 1 - - Reached target Sleep. <30>1 2017-12-18T04:29:23.172199+01:00 Telcontar systemd 1 - - Starting Hibernate... <7>1 2017-12-18T04:29:23.179861+01:00 Telcontar kernel - - - [569194.797529] PM: Hibernation mode set to 'platform' <30>1 2017-12-18T04:29:23.191726+01:00 Telcontar systemd-sleep 16509 - - INFO: running /usr/lib/systemd/system-sleep/grub2.sleep for hibernate <30>1 2017-12-18T04:29:23.191973+01:00 Telcontar systemd-sleep 16509 - - INFO: Running prepare-grub .. <30>1 2017-12-18T04:29:23.192395+01:00 Telcontar systemd-sleep 16509 - - 2017-12-18 04:29:23+01:00 - Hibernating the system now... <30>1 2017-12-18T04:29:23.198353+01:00 Telcontar systemd 1 - - Stopping Deferred execution scheduler... <28>1 2017-12-18T04:29:23.202597+01:00 Telcontar systemd-sh - - - Hibernating the system now... <30>1 2017-12-18T04:29:23.219928+01:00 Telcontar systemd-sleep 16509 - - service: no such service upsd.service <30>1 2017-12-18T04:29:23.220710+01:00 Telcontar systemd 1 - - Stopped Deferred execution scheduler. <30>1 2017-12-18T04:29:23.512897+01:00 Telcontar systemd-sleep 16509 - - running kernel is grub menu entry openSUSE Leap 42.2 (vmlinuz-4.4.92-18.36-default) <30>1 2017-12-18T04:29:23.513124+01:00 Telcontar systemd-sleep 16509 - - preparing boot-loader: selecting entry openSUSE Leap 42.2, kernel /boot/4.4.92-18.36-default <30>1 2017-12-18T04:29:29.227808+01:00 Telcontar systemd-sleep 16509 - - running /usr/sbin/grub2-once "openSUSE Leap 42.2" <30>1 2017-12-18T04:29:29.371469+01:00 Telcontar systemd-sleep 16509 - - time needed for sync: 5.7 seconds, time needed for grub: 0.1 seconds. <30>1 2017-12-18T04:29:29.371708+01:00 Telcontar systemd-sleep 16509 - - INFO: Done. <30>1 2017-12-18T04:29:29.371854+01:00 Telcontar systemd-sleep 16509 - - Suspending system... <6>1 2017-12-18T04:29:41.548570+01:00 Telcontar kernel - - - [569202.198976] PM: Syncing filesystems ... done. <6>1 2017-12-18T04:29:41.747276+01:00 Telcontar kernel - - - [569202.767822] Freezing user space processes ... (elapsed 0.016 seconds) done. <7>1 2017-12-18T04:29:41.747281+01:00 Telcontar kernel - - - [569202.784257] PM: Marking nosave pages: [mem 0x00000000-0x00000fff] <7>1 2017-12-18T04:29:41.747282+01:00 Telcontar kernel - - - [569202.784259] PM: Marking nosave pages: [mem 0x0009f000-0x000fffff] <7>1 2017-12-18T04:29:41.747283+01:00 Telcontar kernel - - - [569202.784262] PM: Marking nosave pages: [mem 0xbff90000-0xffffffff] <7>1 2017-12-18T04:29:41.747284+01:00 Telcontar kernel - - - [569202.784866] PM: Basic memory bitmaps created <6>1 2017-12-18T04:29:41.747286+01:00 Telcontar kernel - - - [569202.784888] PM: Preallocating image memory... <7>1 2017-12-18T04:29:41.747287+01:00 Telcontar kernel - - - [569212.924919] PM: Basic memory bitmaps freed <6>1 2017-12-18T04:29:41.747288+01:00 Telcontar kernel - - - [569212.924928] Restarting tasks ... done. <28>1 2017-12-18T04:29:41.820886+01:00 Telcontar rtkit-daemon 4543 - - The canary thread is apparently starving. Taking action. <29>1 2017-12-18T04:29:41.940237+01:00 Telcontar systemd 1 - - systemd-hibernate.service: Main process exited, code=exited, status=1/FAILURE <30>1 2017-12-18T04:29:41.941655+01:00 Telcontar rtkit-daemon 4543 - - Demoting known real-time threads. <27>1 2017-12-18T04:29:41.941831+01:00 Telcontar systemd 1 - - Failed to start Hibernate. <29>1 2017-12-18T04:29:41.950777+01:00 Telcontar rtkit-daemon 4543 - - Successfully demoted thread 4549 of process 4542 (/usr/bin/pulseaudio). <28>1 2017-12-18T04:29:41.950978+01:00 Telcontar systemd 1 - - Dependency failed for Hibernate. <29>1 2017-12-18T04:29:41.951114+01:00 Telcontar rtkit-daemon 4543 - - Successfully demoted thread 4548 of process 4542 (/usr/bin/pulseaudio). <29>1 2017-12-18T04:29:41.964488+01:00 Telcontar rtkit-daemon 4543 - - Successfully demoted thread 4542 of process 4542 (/usr/bin/pulseaudio). <29>1 2017-12-18T04:29:41.964740+01:00 Telcontar rtkit-daemon 4543 - - Demoted 3 threads. <28>1 2017-12-18T04:29:41.964870+01:00 Telcontar console-kit-daemon 4445 - WARNING: Error waiting for native console 12 activation: Resource temporarily unavailable <28>1 2017-12-18T04:29:41.970606+01:00 Telcontar console-kit-daemon 4445 - WARNING: Error waiting for native console 5 activation: Resource temporarily unavailable <30>1 2017-12-18T04:29:42.037062+01:00 Telcontar console-kit-daemon 4445 - console-kit-daemon[4445]: WARNING: Error waiting for native console 12 activation: Resource temporarily unavailable <30>1 2017-12-18T04:29:42.037261+01:00 Telcontar console-kit-daemon 4445 - console-kit-daemon[4445]: WARNING: Error waiting for native console 5 activation: Resource temporarily unavailable <29>1 2017-12-18T04:29:43.069508+01:00 Telcontar systemd 1 - - hibernate.target: Job hibernate.target/start failed with result 'dependency'. <30>1 2017-12-18T04:29:43.069771+01:00 Telcontar systemd 1 - - sleep.target: Unit not needed anymore. Stopping. <38>1 2017-12-18T04:29:43.078001+01:00 Telcontar systemd-logind 1706 - - Operation 'sleep' finished. <29>1 2017-12-18T04:29:43.081947+01:00 Telcontar systemd 1 - - systemd-hibernate.service: Unit entered failed state. <28>1 2017-12-18T04:29:43.082123+01:00 Telcontar systemd 1 - - systemd-hibernate.service: Failed with result 'exit-code'. <30>1 2017-12-18T04:29:43.091305+01:00 Telcontar systemd 1 - - Stopped target Sleep. <86>1 2017-12-18T04:30:01.946295+01:00 Telcontar cron 16841 - - pam_unix(crond:session): session opened for user cer by (uid=0) <86>1 2017-12-18T04:30:01.947130+01:00 Telcontar cron 16840 - - pam_unix(crond:session): session opened for user wwwrun by (uid=0) <86>1 2017-12-18T04:30:01.947316+01:00 Telcontar cron 16839 - - pam_unix(crond:session): session opened for user root by (uid=0)
The reason is not clear:
systemd 1 - - Dependency failed for Hibernate. systemd-hibernate.service: Failed with result 'exit-code'. systemd-hibernate.service: Main process exited, code=exited, status=1/FAILURE
Well, I know that it failed, but why?
Telcontar:~ # systemctl status systemd-hibernate ● systemd-hibernate.service - Hibernate Loaded: loaded (/usr/lib/systemd/system/systemd-hibernate.service; static; vendor preset: disabled) Active: failed (Result: exit-code) since Mon 2017-12-18 04:29:41 CET; 7h ago Docs: man:systemd-suspend.service(8) Process: 16509 ExecStart=/usr/lib/systemd/systemd-sleep hibernate (code=exited, status=1/FAILURE) Main PID: 16509 (code=exited, status=1/FAILURE)
Dec 18 04:29:23 Telcontar systemd-sleep[16509]: running kernel is grub menu entry openSUSE Leap 42.2 (vmlinuz-4.4.92-18.36-default) Dec 18 04:29:23 Telcontar systemd-sleep[16509]: preparing boot-loader: selecting entry openSUSE Leap 42.2, kernel /boot/4.4.92-18.36-default Dec 18 04:29:29 Telcontar systemd-sleep[16509]: running /usr/sbin/grub2-once "openSUSE Leap 42.2" Dec 18 04:29:29 Telcontar systemd-sleep[16509]: time needed for sync: 5.7 seconds, time needed for grub: 0.1 seconds. Dec 18 04:29:29 Telcontar systemd-sleep[16509]: INFO: Done. Dec 18 04:29:29 Telcontar systemd-sleep[16509]: Suspending system... Dec 18 04:29:41 Telcontar systemd[1]: systemd-hibernate.service: Main process exited, code=exited, status=1/FAILURE Dec 18 04:29:41 Telcontar systemd[1]: Failed to start Hibernate. Dec 18 04:29:43 Telcontar systemd[1]: systemd-hibernate.service: Unit entered failed state. Dec 18 04:29:43 Telcontar systemd[1]: systemd-hibernate.service: Failed with result 'exit-code'. Telcontar:~ #
I know that if I try again it will very probably succeed, but I would like to know why it failed this time.
And of course, on the terminal that I called the command to hibernate, I got no error printed, it just returns as usual.
This is the command I use:
Telcontar:~ # cat /usr/local/bin/hibernate #!/bin/bash date --rfc-3339=seconds systemctl hibernate sleep 3 xscreensaver-command -lock Telcontar:~ #
And this was the result:
Telcontar:~ # hibernate 2017-12-18 04:29:23+01:00 xscreensaver-command: activating and locking.
Telcontar:~ #
-- Cheers
Carlos E. R. (from 42.2 x86_64 "Malachite" at Telcontar)
In those cases where hibernation failed: was a VirtualBox-VM running/loaded? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Thursday, 2017-12-21 at 15:57 +0100, Rainer Fiebig wrote:
In those cases where hibernation failed: was a VirtualBox-VM running/loaded?
I don't know. Yesterday it failed again, and again, and again... it was impossible to hibernate. And the kernel oopsed. And yes, this time an instance of vmplayer was running. I didn't notice at the time; I went out, later updated the machine and rebooted it. - -- Cheers, Carlos E. R. (from openSUSE 42.2 x86_64 "Malachite" at Telcontar) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iEYEARECAAYFAlo7zXsACgkQtTMYHG2NR9ULkgCfeR69uT0gSzs9iHgm1El3bEa6 /7kAn3NW1Wqw1GMOszgOrX5bSh773Pqi =Tsdd -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Carlos E. R. wrote:
On Thursday, 2017-12-21 at 15:57 +0100, Rainer Fiebig wrote:
In those cases where hibernation failed: was a VirtualBox-VM running/loaded?
I don't know.
Yesterday it failed again, and again, and again... it was impossible to hibernate. And the kernel oopsed. And yes, this time an instance of vmplayer was running. I didn't notice at the time; I went out, later updated the machine and rebooted it.
-- Cheers, Carlos E. R. (from openSUSE 42.2 x86_64 "Malachite" at Telcontar)
OK, so you use vmplayer and not VirtualBox? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 21/12/17 15:25, Rainer Fiebig wrote:
Carlos E. R. wrote:
On Thursday, 2017-12-21 at 15:57 +0100, Rainer Fiebig wrote:
In those cases where hibernation failed: was a VirtualBox-VM running/loaded?
I don't know.
Yesterday it failed again, and again, and again... it was impossible to hibernate. And the kernel oopsed. And yes, this time an instance of vmplayer was running. I didn't notice at the time; I went out, later updated the machine and rebooted it.
-- Cheers, Carlos E. R. (from openSUSE 42.2 x86_64 "Malachite" at Telcontar)
OK, so you use vmplayer and not VirtualBox?
Just a tid-bit from my experience, and not with hibernation ... If *I* try to shut down the machine, and *I* am running VB, it kills the shutdown attempt. If *someone else* tries to shut down the machine, and *I* am running VB (or vice versa) then VB gets killed, and I lose my session. In other words, VB actively interferes with this stuff, but can only do so if it's the same user. Cheers, Wol -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Wols Lists wrote:
On 21/12/17 15:25, Rainer Fiebig wrote:
Carlos E. R. wrote:
On Thursday, 2017-12-21 at 15:57 +0100, Rainer Fiebig wrote:
In those cases where hibernation failed: was a VirtualBox-VM running/loaded?
I don't know.
Yesterday it failed again, and again, and again... it was impossible to hibernate. And the kernel oopsed. And yes, this time an instance of vmplayer was running. I didn't notice at the time; I went out, later updated the machine and rebooted it.
-- Cheers, Carlos E. R. (from openSUSE 42.2 x86_64 "Malachite" at Telcontar)
OK, so you use vmplayer and not VirtualBox?
Just a tid-bit from my experience, and not with hibernation ...
If *I* try to shut down the machine, and *I* am running VB, it kills the shutdown attempt.
That's OK with me because one probably wants to shut down the VM cleanly or save it before one shuts down the host.
If *someone else* tries to shut down the machine, and *I* am running VB (or vice versa) then VB gets killed, and I lose my session.
I think it's possible to dis-allow shutdown in a multiple user environment. No? ;)
In other words, VB actively interferes with this stuff, but can only do so if it's the same user.
Cheers, Wol
Well, in case you shutdown the host it gives you a warning - so you can shutdown/save the VM cleanly. But this is not necessary with s2disk/s2both as the VM will be put to sleep like any other app and is unchanged after resume. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Thursday, 2017-12-21 at 17:28 +0100, Rainer Fiebig wrote:
Wols Lists wrote:
On 21/12/17 15:25, Rainer Fiebig wrote:
Carlos E. R. wrote:
On Thursday, 2017-12-21 at 15:57 +0100, Rainer Fiebig wrote:
In those cases where hibernation failed: was a VirtualBox-VM running/loaded?
I don't know.
Yesterday it failed again, and again, and again... it was impossible to hibernate. And the kernel oopsed. And yes, this time an instance of vmplayer was running. I didn't notice at the time; I went out, later updated the machine and rebooted it.
OK, so you use vmplayer and not VirtualBox?
Right.
Just a tid-bit from my experience, and not with hibernation ...
If *I* try to shut down the machine, and *I* am running VB, it kills the shutdown attempt.
That's OK with me because one probably wants to shut down the VM cleanly or save it before one shuts down the host.
Yes, as long as it tells me.
If *someone else* tries to shut down the machine, and *I* am running VB (or vice versa) then VB gets killed, and I lose my session.
I think it's possible to dis-allow shutdown in a multiple user environment. No? ;)
Not trivially, no. - -- Cheers, Carlos E. R. (from openSUSE 42.2 x86_64 "Malachite" at Telcontar) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iEYEARECAAYFAlo8AcAACgkQtTMYHG2NR9XpjwCfeBQGBss4CUjBXoo9uZ72JFyM YNsAn0FArieJPRKn9gGBzPx+dZYiobAh =S2dQ -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Le 21/12/2017 à 19:47, Carlos E. R. a écrit :
I think it's possible to dis-allow shutdown in a multiple user environment. No? ;)
Not trivially, no.
with minimal X config, there no way to shut down the computer for an user not on keyboard jdd -- http://dodin.org -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Thursday, 2017-12-21 at 20:09 +0100, jdd@dodin.org wrote:
Le 21/12/2017 à 19:47, Carlos E. R. a écrit :
I think it's possible to dis-allow shutdown in a multiple user environment. No? ;)
Not trivially, no.
with minimal X config, there no way to shut down the computer for an user not on keyboard
Try "poweroff" from ssh. Of course, I'm root, so... I have not tried as user recently, I do not know what it will do. - -- Cheers, Carlos E. R. (from openSUSE 42.2 x86_64 "Malachite" at Telcontar) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iEYEARECAAYFAlo8STMACgkQtTMYHG2NR9XK5ACaA2yQNuxfx54xSagwctykyxnm g2kAn2OpLLnWt0tamLGMC6ygRnlFK8jX =4hno -----END PGP SIGNATURE-----
Carlos E. R. wrote:
On Thursday, 2017-12-21 at 17:28 +0100, Rainer Fiebig wrote:
Wols Lists wrote:
On 21/12/17 15:25, Rainer Fiebig wrote:
Carlos E. R. wrote:
On Thursday, 2017-12-21 at 15:57 +0100, Rainer Fiebig wrote:
In those cases where hibernation failed: was a VirtualBox-VM running/loaded?
I don't know.
Yesterday it failed again, and again, and again... it was impossible to hibernate. And the kernel oopsed. And yes, this time an instance of vmplayer was running. I didn't notice at the time; I went out, later updated the machine and rebooted it.
OK, so you use vmplayer and not VirtualBox?
Right.
OK. When you have loaded a VM could you do a cat /proc/meminfo in a terminal and post the result? You will see something like this: cat /proc/meminfo MemTotal: 11950220 kB MemFree: 7358740 kB MemAvailable: 8947212 kB Buffers: 1468840 kB Cached: 769932 kB SwapCached: 487700 kB Active: 2167672 kB Inactive: 2211548 kB Active(anon): 2024696 kB Inactive(anon): 563616 kB Active(file): 142976 kB Inactive(file): 1647932 kB Unevictable: 32 kB Mlocked: 32 kB SwapTotal: 10713084 kB SwapFree: 9199480 kB Dirty: 12 kB Writeback: 0 kB AnonPages: 1845744 kB Mapped: 443116 kB Shmem: 447864 kB Slab: 114916 kB SReclaimable: 69304 kB SUnreclaim: 45612 kB KernelStack: 7968 kB PageTables: 52160 kB [...]
Just a tid-bit from my experience, and not with hibernation ...
If *I* try to shut down the machine, and *I* am running VB, it kills the shutdown attempt.
That's OK with me because one probably wants to shut down the VM cleanly or save it before one shuts down the host.
Yes, as long as it tells me.
If *someone else* tries to shut down the machine, and *I* am running VB (or vice versa) then VB gets killed, and I lose my session.
I think it's possible to dis-allow shutdown in a multiple user environment. No? ;)
Not trivially, no.
Yast/Security Center/Boot Settings/Shutdown Behaviour of KDM Login Manager: "Only root" or "Nobody" Not too difficult either. ;)
-- Cheers, Carlos E. R. (from openSUSE 42.2 x86_64 "Malachite" at Telcontar)
-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Thursday, 2017-12-21 at 20:51 +0100, Rainer Fiebig wrote:
Carlos E. R. wrote:
On Thursday, 2017-12-21 at 17:28 +0100, Rainer Fiebig wrote:
Wols Lists wrote:
On 21/12/17 15:25, Rainer Fiebig wrote:
Carlos E. R. wrote:
On Thursday, 2017-12-21 at 15:57 +0100, Rainer Fiebig wrote:
> In those cases where hibernation failed: was a VirtualBox-VM > running/loaded?
I don't know.
Yesterday it failed again, and again, and again... it was impossible to hibernate. And the kernel oopsed. And yes, this time an instance of vmplayer was running. I didn't notice at the time; I went out, later updated the machine and rebooted it.
OK, so you use vmplayer and not VirtualBox?
Right.
OK. When you have loaded a VM could you do a
cat /proc/meminfo
in a terminal and post the result?
You will see something like this:
cat /proc/meminfo MemTotal: 11950220 kB MemFree: 7358740 kB MemAvailable: 8947212 kB Buffers: 1468840 kB Cached: 769932 kB SwapCached: 487700 kB Active: 2167672 kB Inactive: 2211548 kB Active(anon): 2024696 kB Inactive(anon): 563616 kB Active(file): 142976 kB Inactive(file): 1647932 kB Unevictable: 32 kB Mlocked: 32 kB SwapTotal: 10713084 kB SwapFree: 9199480 kB Dirty: 12 kB Writeback: 0 kB AnonPages: 1845744 kB Mapped: 443116 kB Shmem: 447864 kB Slab: 114916 kB SReclaimable: 69304 kB SUnreclaim: 45612 kB KernelStack: 7968 kB PageTables: 52160 kB [...]
Sure. What do you find of interest there? cer@Telcontar:~> cat /proc/meminfo MemTotal: 8174484 kB MemFree: 266964 kB MemAvailable: 2266032 kB Buffers: 104516 kB Cached: 2240604 kB SwapCached: 54736 kB Active: 4881820 kB Inactive: 1699928 kB Active(anon): 4126340 kB Inactive(anon): 1030784 kB Active(file): 755480 kB Inactive(file): 669144 kB Unevictable: 112 kB Mlocked: 112 kB SwapTotal: 25165820 kB SwapFree: 24965896 kB Dirty: 1576 kB Writeback: 0 kB AnonPages: 4183336 kB Mapped: 1221732 kB Shmem: 920496 kB Slab: 981928 kB SReclaimable: 880416 kB SUnreclaim: 101512 kB KernelStack: 14464 kB PageTables: 56432 kB NFS_Unstable: 0 kB Bounce: 0 kB WritebackTmp: 0 kB CommitLimit: 29253060 kB Committed_AS: 10625024 kB VmallocTotal: 34359738367 kB VmallocUsed: 0 kB VmallocChunk: 0 kB HardwareCorrupted: 0 kB AnonHugePages: 1120256 kB HugePages_Total: 0 HugePages_Free: 0 HugePages_Rsvd: 0 HugePages_Surp: 0 Hugepagesize: 2048 kB DirectMap4k: 443968 kB DirectMap2M: 7944192 kB cer@Telcontar:~>
Just a tid-bit from my experience, and not with hibernation ...
If *I* try to shut down the machine, and *I* am running VB, it kills the shutdown attempt.
That's OK with me because one probably wants to shut down the VM cleanly or save it before one shuts down the host.
Yes, as long as it tells me.
If *someone else* tries to shut down the machine, and *I* am running VB (or vice versa) then VB gets killed, and I lose my session.
I think it's possible to dis-allow shutdown in a multiple user environment. No? ;)
Not trivially, no.
Yast/Security Center/Boot Settings/Shutdown Behaviour of KDM Login Manager: "Only root" or "Nobody"
Not too difficult either. ;)
Ah, you mean that. Doesn't work here, because I always powerdown by issuing "poweroff" as root. Besides that, I do not use KDM. Similarly, I hibernate as root by running "systemctl hibernate". Only that I do that from a script I called "hibernate", and I have just inserted in it a test for vmware player so that I will not be caught that way again: #!/bin/bash . /root/ThingsNeededForBoot/MyScriptDefinitions # Copied from old /etc/pm/sleep.d/05vmwarecheck ENUSO=`/etc/init.d/vmware status | grep vmware-vmx` if test -n "$ENUSO" ; then RETVAL=1 echo "Aborting hibernation process because there is a virtual machine running. Stop or suspend it first. " #play -q /root/ThingsNeededForBoot/abitsilly.wav dialgo "Hay una máquina virtual en ejecución todavía." killall gst123 exit fi date --rfc-3339=seconds systemctl hibernate sleep 3 xscreensaver-command -lock - -- Cheers, Carlos E. R. (from openSUSE 42.2 x86_64 "Malachite" at Telcontar) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iEYEARECAAYFAlo8SN0ACgkQtTMYHG2NR9XorwCeJ+Y7Q9MKg0ESHOXSQbjzYJUc MEgAnj0gN7T4FLwAlZHB5PDpPzHPMXyW =ZoP6 -----END PGP SIGNATURE-----
Carlos E. R. wrote:
On Thursday, 2017-12-21 at 20:51 +0100, Rainer Fiebig wrote:
Carlos E. R. wrote:
On Thursday, 2017-12-21 at 17:28 +0100, Rainer Fiebig wrote:
Wols Lists wrote:
On 21/12/17 15:25, Rainer Fiebig wrote:
Carlos E. R. wrote: > > > On Thursday, 2017-12-21 at 15:57 +0100, Rainer Fiebig wrote: > >> In those cases where hibernation failed: was a VirtualBox-VM >> running/loaded? > > I don't know. > > Yesterday it failed again, and again, and again... it was > impossible to > hibernate. And the kernel oopsed. And yes, this time an instance of > vmplayer was running. I didn't notice at the time; I went out, later > updated the machine and rebooted it.
OK, so you use vmplayer and not VirtualBox?
Right.
OK. When you have loaded a VM could you do a
cat /proc/meminfo
in a terminal and post the result?
You will see something like this:
cat /proc/meminfo MemTotal: 11950220 kB MemFree: 7358740 kB MemAvailable: 8947212 kB Buffers: 1468840 kB Cached: 769932 kB SwapCached: 487700 kB Active: 2167672 kB Inactive: 2211548 kB Active(anon): 2024696 kB Inactive(anon): 563616 kB Active(file): 142976 kB Inactive(file): 1647932 kB Unevictable: 32 kB Mlocked: 32 kB SwapTotal: 10713084 kB SwapFree: 9199480 kB Dirty: 12 kB Writeback: 0 kB AnonPages: 1845744 kB Mapped: 443116 kB Shmem: 447864 kB Slab: 114916 kB SReclaimable: 69304 kB SUnreclaim: 45612 kB KernelStack: 7968 kB PageTables: 52160 kB [...]
Sure. What do you find of interest there?
cer@Telcontar:~> cat /proc/meminfo MemTotal: 8174484 kB MemFree: 266964 kB MemAvailable: 2266032 kB Buffers: 104516 kB Cached: 2240604 kB SwapCached: 54736 kB Active: 4881820 kB Inactive: 1699928 kB Active(anon): 4126340 kB Inactive(anon): 1030784 kB Active(file): 755480 kB Inactive(file): 669144 kB Unevictable: 112 kB Mlocked: 112 kB SwapTotal: 25165820 kB SwapFree: 24965896 kB Dirty: 1576 kB Writeback: 0 kB AnonPages: 4183336 kB
This value:
Mapped: 1221732 kB
See https://bugzilla.kernel.org/show_bug.cgi?id=97201 I've sent a patch to the pm-mailing-list. Let's see what happens. Thanks! Rainer Fiebig
Shmem: 920496 kB Slab: 981928 kB SReclaimable: 880416 kB SUnreclaim: 101512 kB KernelStack: 14464 kB PageTables: 56432 kB NFS_Unstable: 0 kB Bounce: 0 kB WritebackTmp: 0 kB CommitLimit: 29253060 kB Committed_AS: 10625024 kB VmallocTotal: 34359738367 kB VmallocUsed: 0 kB VmallocChunk: 0 kB HardwareCorrupted: 0 kB AnonHugePages: 1120256 kB HugePages_Total: 0 HugePages_Free: 0 HugePages_Rsvd: 0 HugePages_Surp: 0 Hugepagesize: 2048 kB DirectMap4k: 443968 kB DirectMap2M: 7944192 kB cer@Telcontar:~>
Just a tid-bit from my experience, and not with hibernation ...
If *I* try to shut down the machine, and *I* am running VB, it kills the shutdown attempt.
That's OK with me because one probably wants to shut down the VM cleanly or save it before one shuts down the host.
Yes, as long as it tells me.
If *someone else* tries to shut down the machine, and *I* am running VB (or vice versa) then VB gets killed, and I lose my session.
I think it's possible to dis-allow shutdown in a multiple user environment. No? ;)
Not trivially, no.
Yast/Security Center/Boot Settings/Shutdown Behaviour of KDM Login Manager: "Only root" or "Nobody"
Not too difficult either. ;)
Ah, you mean that.
Doesn't work here, because I always powerdown by issuing "poweroff" as root.
Besides that, I do not use KDM. Similarly, I hibernate as root by running "systemctl hibernate". Only that I do that from a script I called "hibernate", and I have just inserted in it a test for vmware player so that I will not be caught that way again:
#!/bin/bash
. /root/ThingsNeededForBoot/MyScriptDefinitions
# Copied from old /etc/pm/sleep.d/05vmwarecheck ENUSO=`/etc/init.d/vmware status | grep vmware-vmx` if test -n "$ENUSO" ; then RETVAL=1 echo "Aborting hibernation process because there is a virtual machine running. Stop or suspend it first. " #play -q /root/ThingsNeededForBoot/abitsilly.wav dialgo "Hay una máquina virtual en ejecución todavía." killall gst123 exit fi
date --rfc-3339=seconds systemctl hibernate sleep 3 xscreensaver-command -lock
-- Cheers, Carlos E. R. (from openSUSE 42.2 x86_64 "Malachite" at Telcontar)
-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Friday, 2017-12-22 at 11:36 +0100, Rainer Fiebig wrote:
Carlos E. R. wrote:
On Thursday, 2017-12-21 at 20:51 +0100, Rainer Fiebig wrote:
Carlos E. R. wrote:
On Thursday, 2017-12-21 at 17:28 +0100, Rainer Fiebig wrote:
Wols Lists wrote:
...
OK. When you have loaded a VM could you do a
cat /proc/meminfo
in a terminal and post the result?
...
Sure. What do you find of interest there?
cer@Telcontar:~> cat /proc/meminfo MemTotal: 8174484 kB MemFree: 266964 kB MemAvailable: 2266032 kB Buffers: 104516 kB Cached: 2240604 kB SwapCached: 54736 kB Active: 4881820 kB Inactive: 1699928 kB Active(anon): 4126340 kB Inactive(anon): 1030784 kB Active(file): 755480 kB Inactive(file): 669144 kB Unevictable: 112 kB Mlocked: 112 kB SwapTotal: 25165820 kB SwapFree: 24965896 kB Dirty: 1576 kB Writeback: 0 kB AnonPages: 4183336 kB
This value:
Mapped: 1221732 kB
See https://bugzilla.kernel.org/show_bug.cgi?id=97201
I've sent a patch to the pm-mailing-list. Let's see what happens.
Thanks!
Welcome, but openSUSE does not use pm utils currently, so I'm unsure what that could do for us :-? I noticed this: +++------- Manuel Krause 2017-07-03 17:41:10 UTC I've forgotton to add another comparison: In normal -- unpatched -- mainline 4.11.8 -- KDE & Firefox can remain unresponsive for more than 7mins in corner cases. - -------++- I often noticed the system unresponsive on return from hibernation, specially big apps like firefox. My guess was fragmentation of swap; indeed, I installed an SSD and moved my swap to it: problem solved. - -- Cheers, Carlos E. R. (from openSUSE 42.2 x86_64 "Malachite" at Telcontar) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iEYEARECAAYFAlo9AuoACgkQtTMYHG2NR9X0oACdGBftbHw5pWO6MyfUkMXsJsBV inQAniCcpzQzbHdQZKL/n8A27LDd+wun =Fezg -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Carlos E. R. wrote:
On Friday, 2017-12-22 at 11:36 +0100, Rainer Fiebig wrote:
Carlos E. R. wrote:
On Thursday, 2017-12-21 at 20:51 +0100, Rainer Fiebig wrote:
Carlos E. R. wrote:
On Thursday, 2017-12-21 at 17:28 +0100, Rainer Fiebig wrote:
Wols Lists wrote:
...
OK. When you have loaded a VM could you do a
cat /proc/meminfo
in a terminal and post the result?
...
Sure. What do you find of interest there?
cer@Telcontar:~> cat /proc/meminfo MemTotal: 8174484 kB MemFree: 266964 kB MemAvailable: 2266032 kB Buffers: 104516 kB Cached: 2240604 kB SwapCached: 54736 kB Active: 4881820 kB Inactive: 1699928 kB Active(anon): 4126340 kB Inactive(anon): 1030784 kB Active(file): 755480 kB Inactive(file): 669144 kB Unevictable: 112 kB Mlocked: 112 kB SwapTotal: 25165820 kB SwapFree: 24965896 kB Dirty: 1576 kB Writeback: 0 kB AnonPages: 4183336 kB
This value:
Mapped: 1221732 kB
I've sent a patch to the pm-mailing-list. Let's see what happens.
Thanks!
Welcome, but openSUSE does not use pm utils currently, so I'm unsure what that could do for us :-?
It doesn't matter what user-space-tools are used: s2disk, pm-hibernate etc. The image that is eventually written to disk is created by kernel-functions. And imo there's a bug in one of those functions that makes hibernating completely unreliable if VirtualBox- and perhaps also VMware-Player-VMs are loaded. So the patch is for the kernel.
I noticed this:
+++------- Manuel Krause 2017-07-03 17:41:10 UTC
I've forgotton to add another comparison: In normal -- unpatched -- mainline 4.11.8 -- KDE & Firefox can remain unresponsive for more than 7mins in corner cases. -------++-
I often noticed the system unresponsive on return from hibernation, specially big apps like firefox. My guess was fragmentation of swap; indeed, I installed an SSD and moved my swap to it: problem solved.
-- Cheers, Carlos E. R. (from openSUSE 42.2 x86_64 "Malachite" at Telcontar)
-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Carlos E. R. wrote:
On Friday, 2017-12-22 at 11:36 +0100, Rainer Fiebig wrote:
Carlos E. R. wrote:
On Thursday, 2017-12-21 at 20:51 +0100, Rainer Fiebig wrote:
Carlos E. R. wrote:
On Thursday, 2017-12-21 at 17:28 +0100, Rainer Fiebig wrote:
Wols Lists wrote:
...
OK. When you have loaded a VM could you do a
cat /proc/meminfo
in a terminal and post the result?
...
Sure. What do you find of interest there?
cer@Telcontar:~> cat /proc/meminfo MemTotal: 8174484 kB MemFree: 266964 kB MemAvailable: 2266032 kB Buffers: 104516 kB Cached: 2240604 kB SwapCached: 54736 kB Active: 4881820 kB Inactive: 1699928 kB Active(anon): 4126340 kB Inactive(anon): 1030784 kB Active(file): 755480 kB Inactive(file): 669144 kB Unevictable: 112 kB Mlocked: 112 kB SwapTotal: 25165820 kB SwapFree: 24965896 kB Dirty: 1576 kB Writeback: 0 kB AnonPages: 4183336 kB
This value:
Mapped: 1221732 kB
I've sent a patch to the pm-mailing-list. Let's see what happens.
Thanks!
Welcome, but openSUSE does not use pm utils currently, so I'm unsure what that could do for us :-?
I noticed this:
+++------- Manuel Krause 2017-07-03 17:41:10 UTC
I've forgotton to add another comparison: In normal -- unpatched -- mainline 4.11.8 -- KDE & Firefox can remain unresponsive for more than 7mins in corner cases. -------++-
I often noticed the system unresponsive on return from hibernation, specially big apps like firefox. My guess was fragmentation of swap; indeed, I installed an SSD and moved my swap to it: problem solved.
The image that's written to disk can be huge - depending on memory-load when hibernating. Re-loading that contents into memory can take a while and during that time system and apps may be somewhat unresponsive. An SSD is of course much faster than a HD - this is the reason why your system returns faster to normal responsiveness in high-load-cases. Not fragmentation. But I doubt that it's a good idea to use an SSD for swap. AFAIK it's writing that wears SSDs out. So long!
-- Cheers, Carlos E. R. (from openSUSE 42.2 x86_64 "Malachite" at Telcontar)
-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Friday, 2017-12-22 at 15:21 +0100, Rainer Fiebig wrote:
Carlos E. R. wrote:
I noticed this:
+++------- Manuel Krause 2017-07-03 17:41:10 UTC
I've forgotton to add another comparison: In normal -- unpatched -- mainline 4.11.8 -- KDE & Firefox can remain unresponsive for more than 7mins in corner cases. -------++-
I often noticed the system unresponsive on return from hibernation, specially big apps like firefox. My guess was fragmentation of swap; indeed, I installed an SSD and moved my swap to it: problem solved.
The image that's written to disk can be huge - depending on memory-load when hibernating. Re-loading that contents into memory can take a while and during that time system and apps may be somewhat unresponsive.
An SSD is of course much faster than a HD - this is the reason why your system returns faster to normal responsiveness in high-load-cases. Not fragmentation.
The hard disk handling the swap was working at about 2 MB/sec for minutes, when it is capable of more than 150 (measured). The SSD is capable of 250 MB/S (just measured it, old motherboard). The speed difference is not that big, but the seek time difference is huge. Hours after restoring from hibernation, going from one worskpace displaying Firefox to another displaying LibreOffice could take a minute or two sometimes, just swapping. Now with the SSD (which is not that fast) it just takes seconds. So yes, IMO there is bad swap fragmentation in Leap. There wasn't in 13.1.
But I doubt that it's a good idea to use an SSD for swap. AFAIK it's writing that wears SSDs out.
So be it... - -- Cheers, Carlos E. R. (from openSUSE 42.2 x86_64 "Malachite" at Telcontar) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iEYEARECAAYFAlo9HvAACgkQtTMYHG2NR9UJ5wCeI+Mi3/KE6qbZja2oA0daxMtl e78An0cqss8EiHZQFVzXYR6BsAp2n9rr =mHIn -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 22/12/17 14:21, Rainer Fiebig wrote:
But I doubt that it's a good idea to use an SSD for swap. AFAIK it's writing that wears SSDs out.
Correct. Just like it's flying that wears aircraft out. However, just like an aircraft, by the time an SSD is worn out, it's probably obsolete. In tests, I think it takes a good two years (if not more) to kill a modern SSD, when you're hammering it 24/7 with a script that is intended to kill it ... In normal usage, it will last "for ever", even when swapping - thrashing - like mad. Cheers, Wol -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Wols Lists wrote:
On 22/12/17 14:21, Rainer Fiebig wrote:
But I doubt that it's a good idea to use an SSD for swap. AFAIK it's writing that wears SSDs out.
Correct. Just like it's flying that wears aircraft out.
However, just like an aircraft, by the time an SSD is worn out, it's probably obsolete. In tests, I think it takes a good two years (if not more) to kill a modern SSD, when you're hammering it 24/7 with a script that is intended to kill it ...
In normal usage, it will last "for ever", even when swapping - thrashing - like mad.
Cheers, Wol
So it seems I'm not up-to-date on that topic. Live and learn. So long! -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Le 21/12/2017 à 17:11, Wols Lists a écrit :
On 21/12/17 15:25, Rainer Fiebig wrote:
If *I* try to shut down the machine, and *I* am running VB, it kills the shutdown attempt.
I just tested on my 16Gb ram / 8 Gb swap machine launch virtualbox, windows 7, begin to create something in the explorer, then going to menu, hibernate. relatively long time waiting, then shutoff switch on an resume goes exactly to my pending windows 7 creation in Virtualbox just before hibernation (and found back after):
free total used free shared buffers cached Mem: 16312248 16137416 174832 202628 892628 12016468 -/+ buffers/cache: 3228320 13083928 Swap: 8384508 0 8384508
new "free" after resume
free total used free shared buffers cached Mem: 16312248 5969184 10343064 199440 23908 2993916 -/+ buffers/cache: 2951360 13360888 Swap: 8384508 212380 8172128
jdd -- http://dodin.org -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Le 21/12/2017 à 18:27, jdd@dodin.org a écrit :
switch on an resume goes exactly to my pending windows 7 creation in Virtualbox
but thunderbird refused to send the mail (smtp serveur error), I had to stop it and restart (thunderbird only) jdd -- http://dodin.org -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
participants (5)
-
Carlos E. R. -
jdd＠dodin.org -
Per Jessen -
Rainer Fiebig -
Wols Lists