[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[ale] [systemd] Boot speed

On Sat, 17 Feb 2018 08:58:22 -0500
leam hall via Ale <ale at ale.org> wrote:

> Possibly separating this discussion out into component parts. We'll
> see how it goes.
> I don't see boot speed as a game changer for systemd, even if it is a
> lot faster. 

If your computer is an entertainment appliance receiving live
broadcasts, a 2 second boot is better than a minute boot. If you're
spinning up hundreds of VMs or containers, the boot time of those
matters. If your boot takes 10 minutes, that's unacceptable.

> If you're booting your desktop then you're probably
> already used to "push the power button, hit the head, grab some
> coffee" routine. If you system isn't up by then maybe there's an
> issue.


> For servers, if you really want uptime, why aren't you redundant?
> Reboot time is again not an issue if the service stays up.

In addition, just because Steve Litt once experimentally got systemd
to boot in 2 seconds doesn't mean that's the normal state of affairs.
Reports I hear on various mailing lists have healthy systemd systems
booting in about 20 seconds, and unhealthy ones taking two minutes.

But back to your initial question about boot speed being a game change:
Boot speed being a game changer is an existential necessity to the
systemd cabal because systemd's raison d'?tre is efficiency, and once
the computer is stably up, the init system has little to no effect on
efficiency. The systemd cabal is forced to wow and praise over the boot
speed, all the while saying "and many other things too", because, of
course, you're right: Few care whether boot takes 30 seconds or a
minute, and all too often it's the runit system that takes 30 seconds.

Steve Litt
January 2018 featured book: Troubleshooting: Why Bother?