Systemd 232 Coming Soon With Numerous New Features

[phoronix]

Phoronix: Systemd 232 Coming Soon With Numerous New Features

Systemd 232 is right around the corner to succeed the systemd 231 release from July...

http://www.phoronix.com/scan.php?page=news_item&px=Systemd-232-Features

[uid313]

What is systemd-mount for?

We already have a mount utility called "mount".

[woprandi]

What is systemd-mount for?

We already have a mount utility called "mount".

http://www.phoronix.com/scan.php?page=news_item&px=Systemd-Mount

[rtfazeberdee]

What is systemd-mount for?

We already have a mount utility called "mount".

These might help..

http://www.theregister.co.uk/2016/08/22/systemd_adds_filesystem_mount_tool/

https://www.phoronix.com/scan.php?page=news_item&px=Systemd-Mount

and from the horse's mouth in response to phoronix

https://www.reddit.com/r/linux/comments/4ypb2g/systemd_rolls_out_its_own_mount_tool/d6qicwi

[unixfan2001]

What is systemd-mount for?

We already have a mount utility called "mount".

Sounds like something a Windows user/developer would say.

"Why include OpenSSH? People can just as well download Putty"...

[DanL]

Sometimes, I'm an a$$

[sdack]

systemd-journal-gatewayd-insanid-comingtoyoud-soond
journalctrl-authd-indentid-restrictiond-delayd
lasttod-logoutd-pleased-turnoffd-thelightd
...
And in other news... "less wasn't more after all, says cat to tail"

[Ericg]

What is systemd-mount for?

We already have a mount utility called "mount".

From Lennart

A few clarifications, given that the phoronix article doesnt go into much detail about the background of this:

first of all, this doesn't replace util-linux' mount tool. Not at all. It just tells systemd to mount something, going through systemd's dependency logic. For the actual mount operation PID 1 will fork off util-linux' mount tool like it always did.
systemd-mount is for mount/automount units what systemd-run is for service/scope/timer units: a simple cmdline tool that can create and enqueue these units transiently from the command line.
in contrast to using the mount tool directly by this you get various benefits pid1 provides such as deps, scheduling and sandboxing applied to the mount binary. Thus it will automatically pull in prefix mounts and similar and is much nicer in particular for file systems that fork off a background process.
while systemd-mount can create any kind of mount/automount unit transiently it is particularly neat to use for removable media: consider an usb stick with a typical fat file system on it. Traditionally linux (or more precisely: udisks) would mount the thing on plug and expect the user to unmount it explicitly through the UI before unplugging it. If the fs wasn't unmounted cleanly before umounting, the fs would possibly become corrupted and stay that way. With the logic built into systemd-mount we can drastically improve on this in two ways: first: instead of actually mounting the thing we can just automount it thus establishing the mount point without actually having to mounts the fs. The fs will be mounted on access and a short idle timeout (of 1s) will ensure that the fs is fully clean in most cases and only dirty in a short time frame around the actual fs access. Second: through systemd's job scheduler we can schedule an fsck invocation before the first access. This means: we can automatically fix up the fs should it end up being uncleanly unplugged after all. Both features together i think are a massive step forward for handling removable media: there's a much higher chance that the file systems stay in a healthy state. And all that fully transparent to apps.

So yeah, this is the background of this. And I'd count the last item as a killer feature - at least if you still care about removable media. Quite frankly it's pretty sad that this kind of handling for removable media is only implemented in the year 2016 when online sharing of files has become a lot more common place than removable-media based file sharing.

Lennart

[Michael_S]

Michael, can you modify the article to indicate that systemd-mount wraps the real mount command and layers systemd features on it, and does not supplant it?

I can't blame uid313 for his (or her) confusion about what systemd-mount does.

I would be annoyed at the systemd team for the name, but a more informative name like systemd-mountwrapper or systemd-mount-helper is just too long and awkward to be useful.

[jrch2k8]

Damn i love it, this is gonna be huge on ARM systems and desktops Of couse