I have the standard 13.1/KDE i586 install of Opensuse. Several computers on my network still run 12.3 and they are booting faster than the one with the 13.1 version. Now I disabled the splash screen during boot and found one difference - the 13.1 version stops for approx 15 seconds on the line “Listening on CUPS Printing Service Sockets.”
I have a network printer installed - the connection is socket://192.168.x.x:9100 and the 192.168.x.x is a fixed IP address.
Any idea why the boot process is stopping at this line - has there anything changed between these versions?
Another thing I noticed - when the server is mounted on this computer and I shut down before unmounting the computer does not finish the shutdown and has to be finished by cutting off the power. But I am not sure how this is with the 12.3 version. I only know that with earlier versions I just shut down and it completed the shutdown. Not any longer.
Cheers
Uli
Hi
Have a look at the systemd tools to analyze the true event (it may or may not be that) since things start on demand, in paralell etc.
systemd-analyze
systemd-analyze blame
systemd-analyze critical-chain
(The following as root user)
journalctl -xb
journalctl -xb --no-pager
Interesting, I don’t get that, but then I’m not using (CUPS) socket activation either. Maybe that has an impact?
Thanks for the replies, deano_ferrari and malcolmlewis
malcolmlewis - these commands have hundreds (thousands) of lines what am I looking for:
uli@linux-top:~> systemd-analyze
Startup finished in 5.531s (kernel) + 56.424s (userspace) = 1min 1.956s
uli@linux-top:~> systemd-analyze blame
4.845s home.mount
3.401s dkms_autoinstaller.service
2.105s postfix.service
1.888s SuSEfirewall2.service
1.874s systemd-tmpfiles-clean.service
1.868s cycle.service
1.791s systemd-udev-root-symlink.service
1.581s systemd-fsck@dev-disk-by\x2did-ata\x2dWDC_WD2500BEVT\x2d22ZCT0_WD\x2dWXEY08CR3626\x2dpart7.service
1.552s kmod-static-nodes.service
1.373s vboxdrv.service
1.243s plymouth-start.service
1.082s vboxadd.service
1.071s systemd-vconsole-setup.service
1.065s SuSEfirewall2_init.service
1.052s ModemManager.service
1.035s vmtoolsd.service
924ms xdm.service
781ms systemd-readahead-replay.service
779ms dev-hugepages.mount
776ms systemd-readahead-collect.service
775ms dev-mqueue.mount
769ms sys-kernel-debug.mount
623ms polkit.service
594ms rpcbind.service
588ms nagios.service
569ms systemd-modules-load.service
569ms systemd-remount-fs.service
513ms rsyslog.service
460ms user@1000.service
451ms apparmor.service
410ms smb.service
406ms wpa_supplicant.service
404ms systemd-readahead-done.service
387ms NetworkManager.service
372ms systemd-logind.service
349ms systemd-backlight@acpi_video0.service
303ms systemd-sysctl.service
272ms console-kit-log-system-start.service
262ms alsa-restore.service
252ms avahi-daemon.service
233ms systemd-random-seed.service
231ms nmb.service
205ms systemd-tmpfiles-setup.service
159ms NetworkManager-dispatcher.service
145ms user@0.service
126ms systemd-tmpfiles-setup-dev.service
83ms systemd-user-sessions.service
76ms udisks2.service
75ms rc-local.service
57ms NetworkManager-wait-online.service
50ms bluetooth.service
49ms rtkit-daemon.service
47ms console-kit-daemon.service
43ms dev-disk-by\x2did-ata\x2dWDC_WD2500BEVT\x2d22ZCT0_WD\x2dWXEY08CR3626\x2dpart5.swap
41ms systemd-udev-trigger.service
37ms plymouth-read-write.service
28ms systemd-update-utmp-runlevel.service
24ms upower.service
22ms systemd-journal-flush.service
21ms systemd-udevd.service
16ms var-lock.mount
12ms systemd-update-utmp.service
11ms var-run.mount
uli@linux-top:~> systemd-analyze critical-chain
The time after the unit is active or started is printed after the "@" character.
The time the unit takes to start is printed after the "+" character.
graphical.target @56.395s
└─multi-user.target @56.394s
└─nagios.service @55.805s +588ms
└─cron.service @55.805s
└─postfix.service @53.699s +2.105s
└─nss-lookup.target @53.697s
└─network.target @53.695s
└─NetworkManager.service @53.248s +387ms
└─SuSEfirewall2_init.service @52.173s +1.065s
└─basic.target @52.012s
└─sockets.target @51.926s
└─dbus.socket @51.926s
└─sysinit.target @17.296s
└─apparmor.service @16.844s +451ms
└─systemd-tmpfiles-setup.service @16.638s +205ms
└─local-fs.target @16.624s
└─var-run-vmblock\x2dfuse.mount @52.377s
└─var-run.mount @4.407s +11ms
└─local-fs-pre.target @4.388s
└─systemd-tmpfiles-setup-dev.service @4.260s +126ms
└─kmod-static-nodes.service @2.475s +1.552s
└─systemd-journald.socket @2.472s
└─-.mount @2.468s
└─system.slice @2.485s
└─-.slice @2.484s
The other codes
journalctl -xb
journalctl -xb --no-pager
are even longer. I could get lines out with grep CUPS or similar
Cheers
Uli
Hi
So in the blame section, home.mount is taking the longest time (in seconds) the critical chain shows the order in importance, but as you can see cups is not the issue… If not using vbox or plymouth, disable the services. In the case of plymouth, remove the 5 plymouth rpms and then run mkinitrd should knock 5-8 seconds from your boot time. Then run blame again and check.
thanks, malcolmlewis, I removed the plymouth and virtual box which I haven’t used for quite some time. Ran mkinitrd and ran systemd-analyze blame but got exactly the same result. Do I have to reboot?
Uli
Yes it is an analysis of the last boot you are looking at with blame
Now it looks like this:
uli@linux-top:~> systemd-analyze blame
3.943s dkms_autoinstaller.service
2.572s home.mount
2.388s SuSEfirewall2_init.service
1.914s systemd-udev-root-symlink.service
1.886s cycle.service
1.565s postfix.service
1.483s ModemManager.service
1.411s SuSEfirewall2.service
1.408s avahi-daemon.service
1.386s wpa_supplicant.service
1.371s rsyslog.service
1.363s systemd-logind.service
1.354s rpcbind.service
1.327s kmod-static-nodes.service
1.204s systemd-vconsole-setup.service
1.074s xdm.service
723ms systemd-readahead-collect.service
723ms systemd-readahead-replay.service
722ms dev-hugepages.mount
719ms dev-mqueue.mount
713ms sys-kernel-debug.mount
706ms systemd-fsck@dev-disk-by\x2did-ata\x2dWDC_WD2500BEVT\x2d22ZCT0_WD\x2dWXEY08CR3626\x2dpart7.service
542ms systemd-modules-load.service
482ms alsa-restore.service
481ms console-kit-log-system-start.service
468ms apparmor.service
423ms nagios.service
400ms systemd-remount-fs.service
358ms smb.service
339ms user@1000.service
320ms vmtoolsd.service
265ms systemd-user-sessions.service
214ms systemd-tmpfiles-setup-dev.service
210ms systemd-random-seed.service
183ms nmb.service
178ms systemd-tmpfiles-setup.service
172ms systemd-sysctl.service
104ms NetworkManager.service
76ms polkit.service
70ms udisks2.service
60ms rc-local.service
54ms console-kit-daemon.service
54ms bluetooth.service
49ms NetworkManager-dispatcher.service
45ms dev-disk-by\x2did-ata\x2dWDC_WD2500BEVT\x2d22ZCT0_WD\x2dWXEY08CR3626\x2dpart5.swap
44ms sys-fs-fuse-connections.mount
36ms NetworkManager-wait-online.service
34ms rtkit-daemon.service
31ms systemd-udev-trigger.service
30ms systemd-readahead-done.service
20ms systemd-hostnamed.service
20ms upower.service
17ms systemd-backlight@acpi_video0.service
14ms systemd-journal-flush.service
13ms systemd-udevd.service
13ms systemd-update-utmp-runlevel.service
10ms var-lock.mount
9ms systemd-update-utmp.service
9ms var-run.mount
…and are you still observing the delay you described in your opening post?
I tried after the last post to reboot twice. After GRUB there are a few lines and then the screen remains dark until the login screen comes. No more what is happening and the green OK. So I cannot really compare - I don’t know why that is but otherwise everything seems to be working. I would have to time the boot. Previously I was watching the scrolling text of the boot process and could see where it stops. So at this stage I have no indication where it stops.
Uli
Hi
The only other service I can see you could disable is vmtoolsd.service
systemctl disable vmtoolsd.service
What is the overall time after a reboot now?
systemd-analyze
I’m guessing you need the samba, rsyslog and nagios services for monitoring and file sharing. This is a desktop or laptop, are you using a wireless connection via Network Manager? What is the hardware and ram? And how does this hardware compare to the other 12.3 systems?
just did a blame again:
uli@linux-top:~> systemd-analyze blame
9.494s SuSEfirewall2_init.service
6.741s postfix.service
6.121s SuSEfirewall2.service
5.508s smb.service
5.244s dkms_autoinstaller.service
2.099s ModemManager.service
1.827s cycle.service
1.659s systemd-udev-root-symlink.service
1.640s systemd-fsck@dev-disk-by\x2did-ata\x2dWDC_WD2500BEVT\x2d22ZCT0_WD\x2dWXEY08CR3626\x2dpart7.service
1.256s kmod-static-nodes.service
1.093s home.mount
933ms systemd-logind.service
856ms polkit.service
823ms NetworkManager.service
789ms systemd-readahead-replay.service
787ms dev-hugepages.mount
787ms systemd-readahead-collect.service
787ms dev-mqueue.mount
779ms sys-kernel-debug.mount
744ms systemd-vconsole-setup.service
708ms systemd-remount-fs.service
617ms xdm.service
OK - missed your last post because i posted the first lines of blame with longer times
No I don’t need smb since I have nothing running Windows. I played around with nagios but don’t really need this either. Nagios and virtual box (now uninstalled) did not run on my other laptop. The other laptop has a faster CPU and 4GB RAM versus 2GB here. I watched the speed of boot with the text scrolling past - approx the same scroll speed except the stop described in the first post of this thread. Network connection is via ethernet cable.
I now disabled vmtoolsd.service
systemd-analyze
Startup finished in 5.386s (kernel) + 1min 3.019s (userspace) = 1min 8.406s
Hi
So the smb and nmb services can be diabled and nagios if it still exists, but that may be as good as it gets with your system with 2GB of ram, I’m also guessing a 5400rpm drive? I’m also gussing your using NetworkManager since it’s a laptop, in saying that if the laptop tends to not move location, then consider switch to ifup via YaST Network settings via dhcp or a static ip, gateway and dns addresses required. Oh and if you see the wpa_supplicant service that can be disabled as well.
Thanks malcolmlewis
I don’t know the details of this (you are right) laptop since I got it from my son with Windows not working. It served me well with linux! Turned nagios of at /sbin/chkconfig. How do you turn off smb and nmb services?
Do you know why I don’t see the scrolling text any more -anything to do with plymouth?
Cheers
Uli
They can be turned off with
systemctl stop nmb smb
then permanently disabled with
systemctl disable nmb smb
Hi
You should be able to turn off via systemctl
systemctl disable nmb.service
systemctl disable smb.service
Did you try the esc key? Else modifying the bootloader options via YaST can enable this, however this will probably increase the boot time a little, using the journalctl command to check specific things is better and the systemd-analyze command.
Yes, I tried the Esc key - no change. Not important however
Thank you very much malcolmlewis and deano_ferrari. Your help is very much appreciated.
Cheers
Uli