Re: Suggestion for systemd and /usr on seperate partition
On Wed 08 Jul 2020 at 02:35:09 (+1000), Andrew McGlashan wrote:
> On 8/7/20 2:11 am, Michael Stone wrote:
> > On Tue, Jul 07, 2020 at 10:45:17AM -0500, David Wright wrote:
> >> On Wed 08 Jul 2020 at 00:41:12 (+1000), Andrew McGlashan wrote:
> >>> On 2/11/14 8:58 am, Elimar Riesebieter wrote:
> >>> > * David Baron <d_baron@012.net.il> [2014-11-01 19:13 +0200]:
> >>> >> On Friday 31 October 2014 13:08:27 Elimar Riesebieter wrote:
> >>> >
> >>> > [...]
> >>> >
> >>> >>> It's your decision. MODULES=most should be okay. BUSYBOX=y is
> >>> >>> essential.
> >>> >>
> >>> >> This is what the install gave me. I have not touched it.
> >>> >> Where do I tell it to mount /usr?
> >>> >
> >>> > No need to. initramfs-tools does it by default. Check dmesg or
> >>> > journal.
> >>>
> >>> Still today, it fails to mount /usr if /usr is a logical volume using lvm2
> >>>
> >>> I worked around that problem with an extra "activate" line in the following file:
> >>>
> >>> /usr/share/initramfs-tools/scripts/local-top/lvm2
> >>>
> >>> activate "/dev/mapper/vg0-usr"
> >>>
> >>> I placed that after the line to activate ROOT....
> >>>
> >>> So, still broken after all this time :(
> >>
> >> Is this link worth a read?
> >>
> >> https://freedesktop.org/wiki/Software/systemd/separate-usr-is-broken/
> >>
> >> BTW the first line of the thread is "completely without starting any flamewars:"
> >
> > The short answer is that there simply isn't a good reason to do this on a modern system, and there is no volunteer to donate the enormous amount of effort required to make
> > something work for which there isn't a good justification for expending that effort.
Agreed, but I think both the OP and the reviver are discussing
legacy-partitioned systems. AFAIK these have remained upgradable
on stable from wheezy¹ through to buster—or are you saying that
that's not true? (IDK: all my primary (buster) systems are
/usr-merged too, being installed afresh.)
> > There should be no flamewar, if someone wants the situation to change they simply need to be
> > the person who puts in all the work.
The reason I included that line was that the page includes:
"It isn't systemd's fault. systemd mostly works fine with /usr on a
separate file system that is not pre-mounted at boot."
That had been true for years before this 2014 thread started.
It would be easy to think that the subject line was suggesting this is
not the case, and for the thread to pick up flames. That's all.
> Just doing dist-upgrade with a perfectly acceptable file system previously is no reason why it should break.
Quite right. You read the Release Notes and follow their advice at each upgrade.
> The mentioned intramfs config file has a strange note about it being "dangerous" to enable activate all logical volumes, why?!?!?!
A reference to the specific file would help. I see no mention here.
> Debian/Devuan are Linux distros that allow for continuous upgrading without re-installing; the fact that MANY systems have previously separated root and /usr and, effectively
> "times have changed" really isn't an acceptable answer.
True; I would expect at least a migration path for each distribution.
> Even systemd doesn't seem to think it should be a problem for those that choose to use systemd.
I'm not clear what the significance of "Even …" is. You are presumably
not running systemd, in view of this:
> >>> This is a system now running Devuan Beowulf btw, but it gets most of it's packages directly from Debian repos.
So what steps did you take since, say, wheezy¹ to compensate for the
changes that have been come about?
¹ IIRC booting wheezy did not require anything from /usr until the
scripts in /etc/init.d/ had mounted /usr as per /etc/fstab.
Cheers,
David.
Reply to: