The Wiert Corner – irregular stream of stuff

Jeroen W. Pluimers on .NET, C#, Delphi, databases, and personal interests

  • My badges

  • Twitter Updates

  • My Flickr Stream

  • Pages

  • All categories

  • Enter your email address to subscribe to this blog and receive notifications of new posts by email.

    Join 1,860 other subscribers

Archive for the ‘systemd’ Category

Viewing the last lines of the postfix log with journalctl (with help from Unix & Linux Stack Exchange)

Posted by jpluimers on 2022/02/24

Two command-lines I use to view my Postfix logs:

  1. journalctl --unit postfix --since "2 days ago"
  2. journalctl --unit postfix --pager-end

Note that neither of these work well with the --follow (or equivalent -f) option, as this will effectively disable the pager (which by default is less).

The second is via [Wayback] systemd – How to see the latest x lines from systemctl service log – Unix & Linux Stack Exchange (which got the--pagerend bit wrong, as it misses a dash and should be --pager-end, but still thanks [Wayback] Daniel Kmak):

Just:

journalctl -u SERVICE_NAME -e

Parameter -e stands for:

-e –pagerend

That’s the one ! Other answers will go through the whole log to get to its end, which can be veeeeery long for large syslogs.

The last bit (by [Wayback] Léo Germond, thanks!) is why I like it most.

Similarly, specifying --since in the first example will not go through the whole log.

Some background information:

Read the rest of this entry »

Posted in *nix, *nix-tools, bash, Development, journalctl and journald, Linux, postfix, Power User, Scripting, Software Development, systemd | Leave a Comment »

How do I restart sshd on my Unix system | StarNet Knowledge Database – PC X, X Windows, X 11 & More – StarNet

Posted by jpluimers on 2021/11/17

[Wayback] How do I restart sshd on my Unix system | StarNet Knowledge Database – PC X, X Windows, X 11 & More – StarNet

RedHat and Fedora Core Linux

/sbin/service sshd restart

Suse linux

/etc/rc.d/sshd restart

Debian/Ubuntu

/etc/init.d/sshd restart

Solaris 9 and below

/etc/init.d/sshd stop
/etc/init.d/sshd start

Solaris 10

svcadm disable ssh
svcadm enable ssh

AIX

stopsrc -s sshd
startsrc -s sshd

HP-UX

/sbin/init.d/secsh stop
/sbin/init.d/secsh start

Note that for opensuse, by now you need this to restart sshd:

/usr/sbin/rcsshd restart

Edit 20211118: some tweets in reaction to this post

–jeroen

Posted in *nix, *nix-tools, Debian, Development, Linux, openSuSE, Power User, RedHat, Scripting, Software Development, SuSE Linux, systemd, Tumbleweed | Leave a Comment »

systemd – How to clear journalctl – Unix & Linux Stack Exchange

Posted by jpluimers on 2019/05/24

Some tips on pruning entries from the systemd journal:

For stock opensuse, this is also relevant, as it seems to allow indefinite growth: [WayBack] systemd – journald Settings likey need your attention

You can view disk usage with this command:

journalctl --disk-usage

–jeroen

Posted in *nix, Linux, Power User, systemd | Leave a Comment »

OpenSuSE: multi-megabyte binary blob `/etc/udev/hwdb.bin` does not need to be versioned with `etckeeper` – via Unix & Linux Stack Exchange

Posted by jpluimers on 2019/04/22

I wondered [WayBacklinux – Why is this a binary multi-megabyte blob /etc/udev/hwdb.bin under /etc? – Unix & Linux Stack Exchange.

All my OpenSuSE systems had the /etc/udev/hwdb.bin file with sizes varying between 7.5 and 10 megabytes with many of the zypper dist-upgrade causing updates.

Luckily the system also contains these files which had valid unit definitions, so the file was generated/updated during boot:

  • /usr/bin/systemd-hwdb
  • /usr/lib/systemd/system/sysinit.target.wants/systemd-hwdb-up‌​date.service
  • /usr/lib/systemd/system/systemd-hwdb-update.service

Thanks [WayBacknwildner for answering that question!

So after installing etckeeper and performing etckeeper init you can immediately remove it from the git repository:

# cd /etc
# git rm --cached /etc/udev/hwdb.bin
rm 'udev/hwdb.bin'
# git commit -m "no need for /etc/udev/hwdb.bin to be under revision control as systemd-hwdb maintains it"

For more background, see [WayBack] How to make Git “forget” about a file that was tracked but is now in .gitignore? – Stack Overflow.

–jeroen

 

Posted in *nix, *nix-tools, etckeeper, Linux, openSuSE, Power User, SuSE Linux, systemd, Tumbleweed | Leave a Comment »

systemd/udev: PredictableNetworkInterfaceNames

Posted by jpluimers on 2017/08/28

This is so cool, but will take some time for lot’s of tooling to become compatible:

Starting with v197 systemd/udev will automatically assign predictable, stable network interface names for all local Ethernet, WLAN and WWAN interfaces. This is a departure from the traditional interface naming scheme (“eth0”, “eth1”, “wlan0”, …), but should fix real problems.

–jeroen

Posted in *nix, *nix-tools, Linux, Power User, systemd | Leave a Comment »

OpenSuSE Tumbleweed – when `halt` doesn’t halt, but CLI+HLT the CPU at the end of the shutdown procedure

Posted by jpluimers on 2017/04/26

When halt is not a real halt but a suspend of the CPU.

When halt is not a real halt but a “disabling” of the CPU.

TL;DR:

Don’t use halt, use poweroff instead.

A while ago I wrote about OpenSuSE 12.x not halting after a halt:

The same holds for more recent OpenSuSE systems, but ESXi would never tell what was going on.

Recently I installed an OpenSuSE Tumbleweed system under VMware Fusion (running on Mac OS X) which indicated “The CPU has been disabled by the guest operating system.”

Log indicates a

Log indicates a “Shutdown” which in fact is a CPU not powered down.

Which — Understanding the message: The CPU has been disabled by the guest operating system (2000542) | VMware KB [WayBack] — means that halt will not power down the VM but perform a CLI + HLT on the CPU. This effectively hangs the CPU even though the console log on the right tells does a real Shutdown.

In the past – even under ESXi – a halt would just power down the system, so based on the above I did more digging and fount this very interesting answer in rhel – What is the difference between these commands for bringing down a Linux server? – Unix & Linux Stack Exchange [WayBack] which comes down to:

  • on a systemd [WayBack] based system commands like halt, reboot, shutdown all invoke systemctl [WayBack] calling for a specific target [WayBack].
  • mapping of targets and commands is as follows (quoted from the answer):
    • systemctl isolate halt.target has the shorthands:
      • shutdown -H now
      • systemctl halt
      • plain unadorned halt
    • systemctl isolate reboot.target has the shorthands:
      • shutdown -r now
      • telinit 6
      • systemctl reboot
      • plain unadorned reboot
    • systemctl isolate poweroff.target has the shorthands:
      • shutdown -P now
      • telinit 0
      • shutdown now
      • systemctl poweroff
      • plain unadorned poweroff
    • systemctl isolate rescue.target has the shorthands:
      • telinit 1
      • systemctl rescue
    • systemctl isolate multi-user.target has the shorthands:
      • telinit 2
      • telinit 3
      • telinit 4
    • systemctl isolate graphical.target has the shorthand:
      • telinit 5

For a SysV [WayBack] init runlevels versus systemd targets see:

The systemd parameters making things a bit confusing, for instance you can do reboot --halt and more of those shown in linux – Are there any good reasons for halting system without cutting power? – Super User [WayBack].

That also explains that halt without a powerdown can be useful: it for instance gives the end-user the opportunity to click the reset button instead of the power button after a halt.

–jeroen

Posted in *nix, Linux, openSuSE, Power User, SuSE Linux, systemd, SysVinit, Tumbleweed | Leave a Comment »