Systemd Rolls Out Its Own Mount Tool

[starshipeleven]

Here we go, systemd haters incoming.
If systemd is so bad, why do all major distributions use it instead of moving on without it ?

Lennart has psychic powers, he mind-tricked them.
Actually I think it's a good thing to bring a solid base software stack to Linux. Just look at all the messy boot tooling implementations like Plymouth or Grub.
Let's hope systemd will deliver solutions for that in future and I bet it will see adoption too.[/QUOTE]Already done (although for EFI only), systemd-boot (previously known as gummiboot before it went under Systemd project umbrella).

https://www.freedesktop.org/wiki/Software/systemd/systemd-boot/

Works only for stuff in EFI partition (i.e. it does not load EFI drivers like say rEFInd does to access ext2-3-4 or btrfs partitions, nor it is RAID/LVM/KitchenSink-aware like GRUB2)

See, Lennart has psichic powers. He read your mind.

[starshipeleven]

Damn, I should really start using "Preview Post" function, reposting with fixed quote

Here we go, systemd haters incoming.
If systemd is so bad, why do all major distributions use it instead of moving on without it ?

Lennart has psychic powers, he mind-tricked them.

Actually I think it's a good thing to bring a solid base software stack to Linux. Just look at all the messy boot tooling implementations like Plymouth or Grub.
Let's hope systemd will deliver solutions for that in future and I bet it will see adoption too.

Already done (although for EFI only), systemd-boot (previously known as gummiboot before it went under Systemd project umbrella).

https://www.freedesktop.org/wiki/Software/systemd/systemd-boot/

Works only for stuff in EFI partition (i.e. it does not load EFI drivers like say rEFInd does to access ext2-3-4 or btrfs partitions, nor it is RAID/LVM/KitchenSink-aware like GRUB2)

See, Lennart has psichic powers. He read your mind.

[whitecat]

the problem is that it started as a modern init system replacement

That is not true.

[filssavi]

That is not true.

well some trivial wayback machine search tells me that far far back (2010) the systemd website[1] linked to a blog post from Lennart himself [2] descibing how systemd came to be, and there it calls talks about systemd (and I quote): "This blog story is long, so even though I can only recommend reading the long story, here's the one sentence summary: we are experimenting with a new init system and it is fun."

so it is quite evident that it started as a new ( and so modern, since i dout the aim was to make it legacy style from the get go) init system replacement (since there cannot be 2 init's on a single system, or at least they cannot run simulaneously) what it became after that is history

[filssavi]

well some trivial wayback machine search tells me that far far back (2010) the systemd website[1] linked to a blog post from Lennart himself [2] descibing how systemd came to be, and there it calls talks about systemd (and I quote): "This blog story is long, so even though I can only recommend reading the long story, here's the one sentence summary: we are experimenting with a new init system and it is fun."

so it is quite evident that it started as a new ( and so modern, since i dout the aim was to make it legacy style from the get go) init system replacement (since there cannot be 2 init's on a single system, or at least they cannot run simulaneously) what it became after that is history

missing links:

[1] https://web.archive.org/web/20100724...ftware/systemd
[2] https://web.archive.org/web/20100528...s/systemd.html

[agaman]

Or write "nofail" in the fstab mount options. What about reading documentation? https://www.freedesktop.org/software...emd.mount.html
nofail With nofail, this mount will be only wanted, not required, by local-fs.target or remote-fs.target. This means that the boot will continue even if this mount point is not mounted successfully.

I don't think that "nofail" will allow to do fsck in background... Without that, if the mount fails (because the drive is not found for example) will allow to continue the init as normal. Without it will boot to a rescue mode.
But that reminds me that you can disable fsck in fstab for a mount point. You only have to set the last number of the line to 0.

[balouba]

That is not true.

and orange is blue. wtf is wrong with people.

[starshipeleven]

I don't think that "nofail" will allow to do fsck in background...

Why not? With "nofail" it just gives the order to mount it, then stops caring about it, the fsck is triggered by the mount order, so it is a child of that.

Systemd, like also OpenRC (and theoretically also Upstart) starts stuff concurrently, so it does not have to wait for a process to finish before going on with other stuff.

But that reminds me that you can disable fsck in fstab for a mount point. You only have to set the last number of the line to 0.

if it has issues it will get fscked anyway, that setting is for the "fsck every X times it was mounted or X months" thing.

[sdack]

Here we go, systemd haters incoming. ...

This should make one think, shouldn't it? But apparently it doesn't, which is what makes me sad. Ignorance can be found on all sides. My distribution uses it, too, and it hasn't given me any advantages. What once looked like a speed up on the boot process and which was the reason for me to accept systemd, has become just as slow as it used to be with init scripts and it sometimes even makes me wait for 1:30 min on a shutdown, because something just didn't communicate properly with systemd and so the system won't shut down. So that is what is new for me when I compare old to new. Any advantage of systemd is being used to create more complexity and with it come more problems and bugs. Could I fix a problem in the init scripts myself in the past am I now ever more dependent on my distribution to fix it, or I have to start messing about in C. I can see how this caters towards RedHat, a global money-making business, which needs to make money to stay alive, but I fear the outcome of a microsoftification of Linux cannot be in all our interests. It is not giving me an actual reason to hate it yet (other than the occasional 1:30 min delay during a shutdown).

[Nille_kungen]

You have read the news wrong, none is removing anything.
This is an additional tool with additional features.

Yes and no because i don't think many distributions will have dual for long.
For systemd distributions i think systemd will replace many GNU tools in a not to far away future, i think you foresee this as well.
Users may be able to install the tools but i don't think a user should need to customize so much and there's not to many non systemd distributions right now.

Lennart has psychic powers, he mind-tricked them.

Since systemd seems to be one software to rule them all it's not any mind reading or trick involved.