Hi. I'm jon.404, a Unix/Linux/Database/Openstack/Kubernetes Administrator, AWS/GCP/Azure Engineer, mathematics enthusiast, and amateur philosopher. This is where I rant about that which upsets me, laugh about that which amuses me, and jabber about that which holds my interest most: *nix.

This...this should never happen

Posted 06-30-2018 at 12:11 PM by rocket357
Updated 06-30-2018 at 12:22 PM by rocket357

user@host:~$ sudo reboot
[sudo] password for user:
Failed to start reboot.target: Connection timed out
See system logs and 'systemctl status reboot.target' for details.

Words cannot express how infuriating and depressing that error message is. Just turn the &$(&%^Y(#@ computer off!

Edit: It gets ever more obnoxious:

root@host:~# systemctl status reboot.target
Failed to get properties: Connection timed out

This kind of stupidity is why I reserve using Linux for work (hard to avoid, since Linux has been so integral to my career), but anything personal that matters runs BSD.

Posted in Uncategorized

Views 1663 Comments 7

« Prev Main Next »

Total Comments 7

Comments

SystemD?

Posted 06-30-2018 at 03:16 PM by frankbell
I was going to ask the same question Frank...

Posted 06-30-2018 at 09:10 PM by rkelsen
Yeah systemd was jacked. Not sure what caused it. Hard reboot "fixed" it.

Posted 07-01-2018 at 12:18 AM by rocket357

	Wouldn't happen on Slackware either. If the insanity ever spreads to Slackware I'll likely be joining you in BSD world. I'd have already moved if OpenBSD base was utf8 clean (or still supported iso8859-1) but it looks like they still have work to do on that regard.
	Posted 07-01-2018 at 03:37 AM by GazL

	I am not a fan of SystemD, primarily for philosophical reasons, but I have used a number of distros with SystemD and must say I've found it reliable. I've used Mageia since v. 4 and found the presence of SystemD completely transparent to the user. Have you checked journalctl or the logs to see if they cast any light on what broke?
	Posted 07-01-2018 at 08:36 PM by frankbell Updated 07-01-2018 at 08:39 PM by frankbell

I don't know *precisely* what happened, but there are a few changes on that machine that might account for it. This machine was installed via pxe/preseed initially to be an Openstack compute node. At some point down the road, I'd dropped it out of the Openstack cluster and installed Minecraft on it. It served as a Minecraft server for some time, and it always bugged me that it had an Openstack name instead of a minecraft-related hostname. So I switched the hostname on it.

This is speculation, unfortunately, as the machine has been repurposed again (it's way too powerful to run Minecraft and the 2-3 other services I have on it, so it is now an infra node in the Openstack cluster and one of the other infra nodes was repurposed for Minecraft (fully re-installed with appropriate hostname/disk layout/services this time haha)), but I think the hostname change and systemd didn't agree, or perhaps an Openstack service I *thought* was disabled was causing issues? IIRC I rebooted a few times prior to the above error messages after those changes, though, so I can't say for certain what caused the issue.

Posted 07-02-2018 at 09:46 AM by rocket357 rocket357 is offline

Updated 07-02-2018 at 09:53 AM by rocket357

As it's systemd, I would take a wild guess that you're not supposed to use sudo/su/root login to reboot...

As I recall, systemctl handles all that... and the reboot/shutdown/poweroff/halt/init/runlevel, etc are all replacements for the sysvinit originals which essentially call systemctl and are only there for compatibility.

(I'm not a fan of systemd either, not for philosophical reasons, but because it's an ill conceived and poorly designed mess)

Posted 07-05-2018 at 06:10 AM by cynwulf cynwulf is offline

Updated 07-05-2018 at 06:11 AM by cynwulf

This...this should *never* happen

Comments

This...this should never happen