[gentoo-user] Initramfs or move /usr to /, oh my...

[gentoo-user] Initramfs or move /usr to /, oh my...

[Tanstaafl]

Creating a new thread for this questions since mine got lost in all of 
the follow-ups...

I would really appreciate a meaningful response to this question (maybe 
I should go ask this on -dev?) - this has the potential to lose me 
forever as a gentoo user (I'm sure none of you are crying over that, but 
*I* am), and I've seen other similar comments... I'm thinking of FreeBSD 
too (and PCBSD for my desktop)...

Anyway...

On 2012-03-17 12:11 AM, Bruce Hill, Jr. 
<daddy@happypenguincomputers.com> wrote:
 > An initramfs which does this is created by
 >> =sys-kernel/genkernel-3.4.25.1 or
 >> =sys-kernel/dracut-017-r1. If you do not want to use these tools, be
 > sure any initramfs you create pre-mounts /usr.

Ok, I have never used genkernel, and have no desire to...

I have no idea what dracut is or how to use it...

I have a remote system that has /usr on a separate partition.

So...

How do I find out if I am actually *using* an initramfs right now (I 
know it is built into the kernel), and

If I am not, how do I do this without using genkernel? Is dracut the 
*only* other option? Is it easy/trivial to set one up manually?

I cannot imagine that gentoo is just going to throw me to the wolves 
like this without providing *in-depth* instructions on how to make sure 
my system will boot after this update, like they did with the 
baselayout-2 update...

Personally, I have no problem with not having a separate /usr any more, 
except that I have 3 remote systems that I manage right now that already 
*have* a separate /usr...

On that note - is it possible, and if so, does anyone have any decent 
detailed How-to's on how I might be able to convert a separate /user to 
one on directly on / on a running system?

Re: [gentoo-user] Initramfs or move /usr to /, oh my...

[felix]

On Sun, Mar 18, 2012 at 01:44:07PM -0400, Tanstaafl wrote:
> Creating a new thread for this questions since mine got lost in all of 
> the follow-ups...
> 
> I would really appreciate a meaningful response to this question (maybe 
> I should go ask this on -dev?) - this has the potential to lose me 
> forever as a gentoo user (I'm sure none of you are crying over that, but 
> *I* am), and I've seen other similar comments... I'm thinking of FreeBSD 
> too (and PCBSD for my desktop)...

I wonder what to do also.  Part of me wonders why in the hell anyone
thinks they need to make such a change, seemingly just for the sake of
change.  Having /root, /boot, and /bin et all distinct from the user
mode /usr, /home, and everything else always seemed to me one of the
genuinely clever bits of Unix.  I understand that things get more
complex, and the idea of a very simple base system are long gone, but
why does that require doing away with the separate partitions?

Maybe I'm just a retro grouch in that respect.  But there are other
concerns.  I had thought of just copying /boot et all into /usr,
adding a grub entry to boot off that partition, and easing into the
brave new world.  But I can't do that.  My /usr is an LVM partition,
and making that bootable is apparently as big as hassle, perhaps more
so, than using dracut or some simpler initramfs.

I began computing back before there were integrated circuits and 8 bit
computers, let alone cell phones with more computing power than the
$10M monsters.  I look forward to the day when my pocket computer
automatically links to the display and keyboard at my desk when I sit
down, or projects its display on the wall and watches my fingers on a
bare desk for keys and I don't have to worry about synching my various
computers or worrying about patent wars.  The days have long passed
when I enjoyed seeing how many instructions I could get on one 80
column punched card (hint: overlap them) or how few instructions it
took to figure out the days in a month (hint: use parity) or spending
days optimizing for a drum computer ... or messing with configuration
issues because some self-proclaimed efficiency export decided that
/usr was needed at boot.

My attitude right now is to wait and see.  Maybe this will all blow
over, maybe the self-proclaimed experts will find other things to do,
maybe other self-proclaimed experts will find nifty tools to make
migration easier.  In the meantime, I have other work to do, and I
will just freeze parts of my system for the time being.  I don't see
migrating to other systems as being worth any more than an up-yours.
Any other linux system will no doubt do the same.  Any other unix but
not linux system will have an entirely different hassle.  I am past
the days when dinking for the sake of dinking involved boot issues and
disk configurations.  There are much more interesting bigger issues to
dink with now.

-- 
            ... _._. ._ ._. . _._. ._. ___ .__ ._. . .__. ._ .. ._.
     Felix Finch: scarecrow repairman & rocket surgeon / felix@crowfix.com
  GPG = E987 4493 C860 246C 3B1E  6477 7838 76E9 182E 8151 ITAR license #4933
I've found a solution to Fermat's Last Theorem but I see I've run out of room o

Re: [gentoo-user] Initramfs or move /usr to /, oh my...

[Jarry]

On 18-Mar-12 18:44, Tanstaafl wrote:

> How do I find out if I am actually *using* an initramfs right now (I
> know it is built into the kernel),

Just guessing: If you do not know, then you are probably not using it...

Jarry

-- 
_______________________________________________________________
This mailbox accepts e-mails only from selected mailing-lists!
Everything else is considered to be spam and therefore deleted.

Re: [gentoo-user] Initramfs or move /usr to /, oh my...

[Alex Schuster]

Tanstaafl writes:

> On 2012-03-17 12:11 AM, Bruce Hill, Jr. 
> <daddy@happypenguincomputers.com> wrote:
>  > An initramfs which does this is created by
>  >> =sys-kernel/genkernel-3.4.25.1 or
>  >> =sys-kernel/dracut-017-r1. If you do not want to use these tools, be
>  > sure any initramfs you create pre-mounts /usr.
> 
> Ok, I have never used genkernel, and have no desire to...

I started using it when I encrypted my whole hard drive, so I needed an
initramfs. It worked just fine. I had to set MENUICONFIG="yes"
and CLEAN="no" in genkernel.conf, if not I think genkernel generates a
new .config which is not what I wanted. genkernel --install --lvm --luks
all was all that is needed then. Yes, I read that you don't want to use
it, but I thought I'd mention it just in case.

> I have no idea what dracut is or how to use it...

I also did not use that yet.

> I have a remote system that has /usr on a separate partition.
> 
> So...
> 
> How do I find out if I am actually *using* an initramfs right now (I 
> know it is built into the kernel), and

I'd say if there is no "initrd" line in you grub.conf, and no
corresponding file in /boot, you don't use one. And you're using Gentoo,
where there is no automatic setup of initramfs stuff, so it is highly
unlikely you are using one without knowing.

> If I am not, how do I do this without using genkernel? Is dracut the 
> *only* other option?

No, but probably the easiest.

> Is it easy/trivial to set one up manually?

Hmm, not really. I did some experiments, but it was too much work
for me, and I decided to use one of the tools (genkernel) that are
available. You'd have to create a gzipped cpio archive containing all the
needed stuff, binaries, libraries, kernel modules, and an init script
which handles everything that needs be done, like mounting /usr. 

> I cannot imagine that gentoo is just going to throw me to the wolves 
> like this without providing *in-depth* instructions on how to make sure 
> my system will boot after this update, like they did with the 
> baselayout-2 update...

I'm also wondering.

> Personally, I have no problem with not having a separate /usr any more, 
> except that I have 3 remote systems that I manage right now that
> already *have* a separate /usr...
> 
> On that note - is it possible, and if so, does anyone have any decent 
> detailed How-to's on how I might be able to convert a separate /user to 
> one on directly on / on a running system?

Is your root partition large enough? Then just copy the stuff over:

  mount -o bind / /mnt # makes / available in /mnt, without other
                       # partitions like /usr showing up there
  cp -a /usr /mnt/

And remove /usr from /etc/fstab before rebooting.

If there's not enough space, you need to enlarge the partition. Very easy
with LVM, but if you were using it on your root file system, you'd
already be using an initramfs. If not, you need to take the machine down
anyway and use gparted or something from a live-cd to adjust your
partitions.

	Wonko

Re: [gentoo-user] Initramfs or move /usr to /, oh my...

[Canek Peláez Valdés]

On Sun, Mar 18, 2012 at 11:44 AM, Tanstaafl <tanstaafl@libertytrek.org> wrote:

[snip]

> Ok, I have never used genkernel, and have no desire to...
>
> I have no idea what dracut is or how to use it...
>
> I have a remote system that has /usr on a separate partition.
>
> So...
>
> How do I find out if I am actually *using* an initramfs right now (I know it
> is built into the kernel), and
>
> If I am not, how do I do this without using genkernel? Is dracut the *only*
> other option? Is it easy/trivial to set one up manually?

udev is going to be unmasked, not stabilized. By the time udev gets
into x86/amd64, hopefully the documentation necessary will be ready.

You can suscribe to bug 407959
(https://bugs.gentoo.org/show_bug.cgi?id=407959), which tracks the
documentation changes necessary. Right now the only blocker is 408691
(https://bugs.gentoo.org/show_bug.cgi?id=408691), but I'm sure it will
be joined by more bugs in the near future.

Devs are already working on the documentation. If you have a test
spare machine, you can help them, and the whole Gentoo comunity.

Regards.
-- 
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México

[gentoo-user] Re: Initramfs or move /usr to /, oh my...

[walt]

On 03/18/2012 10:44 AM, Tanstaafl wrote:
> Creating a new thread for this questions since mine got lost in all
> of the follow-ups...

I actually read a good response, but I can't possibly find it again :)
I do recall that it said to look at your grub.conf (menu.lst) to see
if grub passes an initrd= or rdinit= to the kernel during bootup.

The other nifty hint was to add "panic=10" as a kernel parameter in
grub.conf (menu.lst) so that your remote system will reboot in 10
seconds if the kernel panics during boot.  That will let you test
(remotely) if a kernel parameter like "noinitrd" breaks your machine.

> I have a remote system that has /usr on a separate partition.
> 
> So...
> 
> How do I find out if I am actually *using* an initramfs right now (I
> know it is built into the kernel), and
> 
> If I am not, how do I do this without using genkernel? Is dracut the
> *only* other option? Is it easy/trivial to set one up manually?
> 
> I cannot imagine that gentoo is just going to throw me to the wolves
> like this without providing *in-depth* instructions on how to make
> sure my system will boot after this update, like they did with the
> baselayout-2 update...
> 
> Personally, I have no problem with not having a separate /usr any
> more, except that I have 3 remote systems that I manage right now
> that already *have* a separate /usr...
> 
> On that note - is it possible, and if so, does anyone have any decent
> detailed How-to's on how I might be able to convert a separate /user
> to one on directly on / on a running system?
> 
> 

Re: [gentoo-user] Initramfs or move /usr to /, oh my...

[Dale]

Tanstaafl wrote:
> Creating a new thread for this questions since mine got lost in all of
> the follow-ups...
> 
> I would really appreciate a meaningful response to this question (maybe
> I should go ask this on -dev?) - this has the potential to lose me
> forever as a gentoo user (I'm sure none of you are crying over that, but
> *I* am), and I've seen other similar comments... I'm thinking of FreeBSD
> too (and PCBSD for my desktop)...
> 
> Anyway...
> 
> On 2012-03-17 12:11 AM, Bruce Hill, Jr.
> <daddy@happypenguincomputers.com> wrote:
>> An initramfs which does this is created by
>>> =sys-kernel/genkernel-3.4.25.1 or
>>> =sys-kernel/dracut-017-r1. If you do not want to use these tools, be
>> sure any initramfs you create pre-mounts /usr.
> 
> Ok, I have never used genkernel, and have no desire to...
> 
> I have no idea what dracut is or how to use it...
> 
> I have a remote system that has /usr on a separate partition.
> 
> So...
> 
> How do I find out if I am actually *using* an initramfs right now (I
> know it is built into the kernel), and
> 
> If I am not, how do I do this without using genkernel? Is dracut the
> *only* other option? Is it easy/trivial to set one up manually?
> 
> I cannot imagine that gentoo is just going to throw me to the wolves
> like this without providing *in-depth* instructions on how to make sure
> my system will boot after this update, like they did with the
> baselayout-2 update...
> 
> Personally, I have no problem with not having a separate /usr any more,
> except that I have 3 remote systems that I manage right now that already
> *have* a separate /usr...
> 
> On that note - is it possible, and if so, does anyone have any decent
> detailed How-to's on how I might be able to convert a separate /user to
> one on directly on / on a running system?
> 
> 


I'm going to add this.  I have been using a init thingy that I used
dracut to build.  When I boot using the init thingy, my system doesn't
work right.  I am able to reproduce this too.  Right now, if I use the
init thingy, I can't use part of my system that for me is vital.  I
can't switch from user to root in anything, not even a console.  So right
now, I'm having to boot without the init thingy and still want to
migrate /usr to LVM.  That is certainly not going to happen right now.

My advice, mask udev to what works for you until all this mess get
sorted out.  The first time someone tries to ssh in as a user, then su to
root, they are going to have a bad day if they run into the issue I am
having.  Remember, most admins set remote systems not to allow root to
login directly as a security feature.  That's what I have read anyway.

Just keep it so you can use it until you know the bugs are sorted out.
I'm still trying to figure mine out.

Dale

:-)  :-)


-- 
I am only responsible for what I said ... Not for what you understood or
how you interpreted my words!

Miss the compile output?  Hint:
EMERGE_DEFAULT_OPTS="--quiet-build=n"

[gentoo-user] Re: Initramfs or move /usr to /, oh my...

[walt]

On 03/18/2012 11:52 AM, walt wrote:

> The other nifty hint was to add "panic=10" as a kernel parameter in
> grub.conf (menu.lst) so that your remote system will reboot in 10
> seconds if the kernel panics during boot.  That will let you test
> (remotely) if a kernel parameter like "noinitrd" breaks your machine.

Heh.  I learn a lot from reading my posts -- when I figure out why
my first reply was wrong :p

Now that I've thought about it, I assume you have only ssh access to
your remote machine, so you can't see the grub boot prompt, right?

Maybe the remote machine doesn't even pause at the boot prompt because
no one is there to watch it?  I'm curious how remote servers work in
real life because in my next life I wanna come back as a sysadmin :)

[gentoo-user] Re: Initramfs or move /usr to /, oh my...

[»Q«]

On Sun, 18 Mar 2012 13:44:07 -0400
Tanstaafl <tanstaafl@libertytrek.org> wrote:

> I cannot imagine that gentoo is just going to throw me to the wolves 
> like this without providing *in-depth* instructions on how to make
> sure my system will boot after this update, like they did with the 
> baselayout-2 update...

They plan to provide transition and hope to do better than with the
baselayout-2 transition.  I haven't tried to dig up the posts saying
so, as there is a much noise about all this on the dev list as here.

Re: [gentoo-user] Re: Initramfs or move /usr to /, oh my...

[Pandu Poluan]

[-- Attachment #1: Type: text/plain, Size: 1185 bytes --]

On Mar 19, 2012 5:31 AM, "walt" <w41ter@gmail.com> wrote:
>
> On 03/18/2012 11:52 AM, walt wrote:
>
> > The other nifty hint was to add "panic=10" as a kernel parameter in
> > grub.conf (menu.lst) so that your remote system will reboot in 10
> > seconds if the kernel panics during boot.  That will let you test
> > (remotely) if a kernel parameter like "noinitrd" breaks your machine.
>
> Heh.  I learn a lot from reading my posts -- when I figure out why
> my first reply was wrong :p
>
> Now that I've thought about it, I assume you have only ssh access to
> your remote machine, so you can't see the grub boot prompt, right?
>
> Maybe the remote machine doesn't even pause at the boot prompt because
> no one is there to watch it?  I'm curious how remote servers work in
> real life because in my next life I wanna come back as a sysadmin :)
>

When I started administering remote servers, Citrix's XenServer is Good
Enough™ to deploy in production, so now it's the first thing I install on a
virgin box, even if said virgin box will host only one VM.

This provides me with a usable Virtual Console through which I can watch
the boot process.

Rgds,

[-- Attachment #2: Type: text/html, Size: 1458 bytes --]

Re: [gentoo-user] Initramfs or move /usr to /, oh my...

[Bruce Hill, Jr.]

 On March 18, 2012 at 2:52 PM "Canek Peláez Valdés" <caneko@gmail.com>
wrote:

> On Sun, Mar 18, 2012 at 11:44 AM, Tanstaafl <tanstaafl@libertytrek.org>
wrote:
>
> [snip]
>
> > Ok, I have never used genkernel, and have no desire to...
> >
> > I have no idea what dracut is or how to use it...
> >
> > I have a remote system that has /usr on a separate partition.
> >
> > So...
> >
> > How do I find out if I am actually *using* an initramfs right now (I
know it
> > is built into the kernel), and

ls -l /boot/ will tell you.

There is a difference between an initrd (initial RAM disk) image (simple)
and an initramfs (initial RAM filesystem) (complicated). Gentoo used to
have a script called mkinitrd. It was removed before I migrated to Gentoo,
but I should look in attic to see if it's still there. To date I've found
no one in Gentoo who will even discuss it.

Slackware has used mkinitrd for ions, and it still works very efficiently
there. Of course, Eric Hameleers understands the script, and Slackware's
init scripts, and maintains mkinitrd. Maybe in Gentoo somebody "upstream"
scared people with initramfs, like they're doing with this horrible systemd
idea, and whoever maintained mkinitrd just cowered in the corner and
dropped the ball. Who knows?

The bottom line is that "officially" Gentoo has abandoned initrd for
initramfs. You can write a script to make an initrd, as people do all the
time. But don't look for official Gentoo support for it.

It seems to me after a year around Gentoo that things get so complicated,
and upstream gets to "force" things on Gentoo (such as systemd), because
there are just too many different developers. All are human with different
opinions, so you wind up with people going off in different directions with
no cohesive ability to stand against upstream. IOW, we're too forked within
Gentoo.

For instance, the maintainer of ConsoleKit in Gentoo (Gnome herd guy) says
he doesn't care about systemd, he's maintaining ConsoleKit and it's not
going anywhere. (We'll see...)

Anyway ... for more on Gentoo's initramfs read
http://en.gentoo-wiki.com/wiki/Initramfs

> > If I am not, how do I do this without using genkernel? Is dracut the
*only*
> > other option? Is it easy/trivial to set one up manually?

> 
> On March 18, 2012 at 2:52 PM "Canek Peláez Valdés" <caneko@gmail.com>
prattled:
> 
> udev is going to be unmasked, not stabilized. By the time udev gets
> into x86/amd64, hopefully the documentation necessary will be ready.

That's you telling the world what an asinine idea this drastic change is
... when it's the stable version, which most of the unsuspecting Gentoo
userbase will emerge, "hopefully the documentation necessary will be
ready." Par for the Poettering course.

> You can suscribe to bug 407959
> (https://bugs.gentoo.org/show_bug.cgi?id=407959), which tracks the
> documentation changes necessary. Right now the only blocker is 408691
> (https://bugs.gentoo.org/show_bug.cgi?id=408691), but I'm sure it will
> be joined by more bugs in the near future.
>
> Devs are already working on the documentation. If you have a test
> spare machine, you can help them, and the whole Gentoo comunity.
>
> Regards.
> --
> Canek Peláez Valdés

udev is already unmasked, and stabilized at 171-r5 now...
You need to explain what you mean ... you're probably talking about
udev-181

Please don't encourage people who don't understand what's happening to test
nefarious software ideas. There is nothing about this that's going to "help
the whole Gentoo community."
--
Happy Penguin Computers    >`)
126 Fenco Drive            ( \
Tupelo, MS 38801            ^^
662-269-2706; 662-491-8613
support at happypenguincomputers dot com
http://www.happypenguincomputers.com

[gentoo-user] Re: Initramfs or move /usr to /, oh my...

[walt]

On 03/18/2012 04:31 PM, Pandu Poluan wrote:
> 
> On Mar 19, 2012 5:31 AM, "walt" <w41ter@gmail.com
> <mailto:w41ter@gmail.com>> wrote:
>> 
>> On 03/18/2012 11:52 AM, walt wrote:
>> 
>>> The other nifty hint was to add "panic=10" as a kernel parameter
>>> in grub.conf (menu.lst) so that your remote system will reboot in
>>> 10 seconds if the kernel panics during boot.  That will let you
>>> test (remotely) if a kernel parameter like "noinitrd" breaks your
>>> machine.
>> 
>> Heh.  I learn a lot from reading my posts -- when I figure out why 
>> my first reply was wrong :p
>> 
>> Now that I've thought about it, I assume you have only ssh access
>> to your remote machine, so you can't see the grub boot prompt,
>> right?
>> 
>> Maybe the remote machine doesn't even pause at the boot prompt
>> because no one is there to watch it?  I'm curious how remote
>> servers work in real life because in my next life I wanna come back
>> as a sysadmin :)
>> 
> 
> When I started administering remote servers, Citrix's XenServer is
> Good Enough™ to deploy in production, so now it's the first thing I
> install on a virgin box, even if said virgin box will host only one
> VM.
> 
> This provides me with a usable Virtual Console through which I can
> watch the boot process.

Bless you Pandu, you just answered a question I didn't ask (yet :)

My workplace recently began providing us (peons) access to its Holy
Intranet even when we are (shamefully) not actually in the workplace.

When I use firefox to access their intranet I have no problems: I
see a small popup dialog box that announces that Citrix is allowing
me to see a window containing an instance of M$ Internet Explorer,
which is displaying the intranet web page I clicked on in firefox,
(which is running on my gentoo desktop, of course).

I can see that this whole process starts a java vm running in the
background, so I suppose that the Citrix app (whatever it is) is
a java applet started by my firefox browser.

But, when I try to access the same intranet web page with google
chrome, it hangs forever instead of starting the Citrix app.
(Other java-powered websites work normally with google chrome.)

Does any/all of this suggest that their web servers are running the
same Citrix XenServer you speak of?

Re: [gentoo-user] Re: Initramfs or move /usr to /, oh my...

[Pandu Poluan]

[-- Attachment #1: Type: text/plain, Size: 3171 bytes --]

On Mar 19, 2012 9:41 AM, "walt" <w41ter@gmail.com> wrote:
>
> On 03/18/2012 04:31 PM, Pandu Poluan wrote:
> >
> > On Mar 19, 2012 5:31 AM, "walt" <w41ter@gmail.com
> > <mailto:w41ter@gmail.com>> wrote:
> >>
> >> On 03/18/2012 11:52 AM, walt wrote:
> >>
> >>> The other nifty hint was to add "panic=10" as a kernel parameter
> >>> in grub.conf (menu.lst) so that your remote system will reboot in
> >>> 10 seconds if the kernel panics during boot.  That will let you
> >>> test (remotely) if a kernel parameter like "noinitrd" breaks your
> >>> machine.
> >>
> >> Heh.  I learn a lot from reading my posts -- when I figure out why
> >> my first reply was wrong :p
> >>
> >> Now that I've thought about it, I assume you have only ssh access
> >> to your remote machine, so you can't see the grub boot prompt,
> >> right?
> >>
> >> Maybe the remote machine doesn't even pause at the boot prompt
> >> because no one is there to watch it?  I'm curious how remote
> >> servers work in real life because in my next life I wanna come back
> >> as a sysadmin :)
> >>
> >
> > When I started administering remote servers, Citrix's XenServer is
> > Good Enough™ to deploy in production, so now it's the first thing I
> > install on a virgin box, even if said virgin box will host only one
> > VM.
> >
> > This provides me with a usable Virtual Console through which I can
> > watch the boot process.
>
> Bless you Pandu, you just answered a question I didn't ask (yet :)
>
> My workplace recently began providing us (peons) access to its Holy
> Intranet even when we are (shamefully) not actually in the workplace.
>
> When I use firefox to access their intranet I have no problems: I
> see a small popup dialog box that announces that Citrix is allowing
> me to see a window containing an instance of M$ Internet Explorer,
> which is displaying the intranet web page I clicked on in firefox,
> (which is running on my gentoo desktop, of course).
>
> I can see that this whole process starts a java vm running in the
> background, so I suppose that the Citrix app (whatever it is) is
> a java applet started by my firefox browser.
>
> But, when I try to access the same intranet web page with google
> chrome, it hangs forever instead of starting the Citrix app.
> (Other java-powered websites work normally with google chrome.)
>
> Does any/all of this suggest that their web servers are running the
> same Citrix XenServer you speak of?
>
>

That's XenApp in action, and despite having "Xen" in its name, it's not
Xen. IOW, that's application virtualization not baremetal OS Virtualization.

You won't ever know if your server is running XenServer or not, unless you
ask the SysAdmin team. Yes, it's that transparent.

XenApp is a love letter from hell, if you ask me. Here in my new employer,
it's used extensively to access apps that actually live in the Windows
Servers. The (quite sizable) dev team is currently feverishly trying to
finish proper client/server apps to replace them. Reason for replacement?
Well, your experience is but one of the teeth-gritting problems we're
experiencing :-P

Rgds,

[-- Attachment #2: Type: text/html, Size: 4027 bytes --]

Re: [gentoo-user] Initramfs or move /usr to /, oh my...

[Neil Bothwick]

[-- Attachment #1: Type: text/plain, Size: 791 bytes --]

On Sun, 18 Mar 2012 21:57:39 -0400 (EDT), Bruce Hill, Jr. wrote:

> > > How do I find out if I am actually *using* an initramfs right now
> > > (I  know it
> > > is built into the kernel), and  
> 
> ls -l /boot/ will tell you.


No it won't, as the initramfs can be built into the kernel image.
 
> There is a difference between an initrd (initial RAM disk) image
> (simple) and an initramfs (initial RAM filesystem) (complicated).

I think you have that the wrong way round. An initramfs is simpler in
that you don't have to worry about creating a filesystem or even creating
the image file, just give the kernel compilation process a file
containing a list of what should be in it.


-- 
Neil Bothwick

Middle-age - because your age starts to show at your middle.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [gentoo-user] Initramfs or move /usr to /, oh my...

[Sebastian Pipping]

On 03/18/2012 06:44 PM, Tanstaafl wrote:
> Ok, I have never used genkernel, and have no desire to...
> 
> I have no idea what dracut is or how to use it...

While genkernel also can generate kernel configs for you, both dracut
and genkernel are initramfs creators: they take repeated creation (after
each kernel update) of a potentially complex initramfs off your shoulders.


> I have a remote system that has /usr on a separate partition.
> 
> So...
> 
> How do I find out if I am actually *using* an initramfs right now (I
> know it is built into the kernel)

Afaik the in-kernel one is a dummy, only.  If you don't remember
anything about genkernel or dracut from the past you do not have an
initramfs in your Gentoo installation(s).


>, and
> 
> If I am not, how do I do this without using genkernel? Is dracut the
> *only* other option? Is it easy/trivial to set one up manually?

A manual initramfs is not that easy, no.  And it becomes outdated more
easily than an initramfs-creator-based approach.


> On that note - is it possible, and if so, does anyone have any decent
> detailed How-to's on how I might be able to convert a separate /user to
> one on directly on / on a running system?

From my current understanding (please double-check, no warrenties):

 0. Make backups

 1. Boot some sort of live/rescue CD
    (so you can fiddle with /usr without shooting your foot)

(2. Enlarge space on partition/device <root> (the one holding /))

(3. Enlarge file system sitting on partition/device <root>)

 4. Make a new folder <root>/usr

 5. Copy content from <usr>/ to <root>/usr/

    - Watch out for use of Xattr (extended file attributes)

    - Watch out for use of POSIX ACLs

    - Use something like --archive with cp/rsync to maintain attributes

 6. Update <root>/etc/fstab

 7. Reboot

 8. Resolve partition <usr>

Good luck.

Best,



Sebastian

Re: [gentoo-user] Initramfs or move /usr to /, oh my...

[Neil Bothwick]

[-- Attachment #1: Type: text/plain, Size: 506 bytes --]

On Tue, 20 Mar 2012 02:28:20 +0100, Sebastian Pipping wrote:

> A manual initramfs is not that easy, no.  And it becomes outdated more
> easily than an initramfs-creator-based approach.

If you keep a separate directory for your initramfs files, this is true.
But if your initramfs config is a list of files to include, the current
versions of everything will always be included.


-- 
Neil Bothwick

"If you can't explain it simply, you don't understand it well enough."
 (Albert Einstein)

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [gentoo-user] Initramfs or move /usr to /, oh my...

[Sebastian Pipping]

On 03/20/2012 09:43 AM, Neil Bothwick wrote:
> If you keep a separate directory for your initramfs files, this is true.
> But if your initramfs config is a list of files to include, the current
> versions of everything will always be included.

Fair point.

Best,



Sebastian

Re: [gentoo-user] Initramfs or move /usr to /, oh my...

[Joshua Murphy]

On Mon, Mar 19, 2012 at 9:28 PM, Sebastian Pipping <sping@gentoo.org> wrote:
> On 03/18/2012 06:44 PM, Tanstaafl wrote:
<snip>
>> On that note - is it possible, and if so, does anyone have any decent
>> detailed How-to's on how I might be able to convert a separate /user to
>> one on directly on / on a running system?
>
> From my current understanding (please double-check, no warrenties):
>
>  0. Make backups
>
>  1. Boot some sort of live/rescue CD
>    (so you can fiddle with /usr without shooting your foot)
>
> (2. Enlarge space on partition/device <root> (the one holding /))
>
> (3. Enlarge file system sitting on partition/device <root>)
>
>  4. Make a new folder <root>/usr
>
>  5. Copy content from <usr>/ to <root>/usr/
>
>    - Watch out for use of Xattr (extended file attributes)
>
>    - Watch out for use of POSIX ACLs
>
>    - Use something like --archive with cp/rsync to maintain attributes
>
>  6. Update <root>/etc/fstab
>
>  7. Reboot
>
>  8. Resolve partition <usr>
>
> Good luck.
>
> Best,
>
>
>
> Sebastian
>

That would work, and is among the safer routes, but it also means
taking the system down for a fair while. If your root has enough free
space to hold all of /usr already, without any resizing, you actually
can do everything with only a single reboot along the way (and while
that, too, *can* be avoided with a lazy unmount in place of the reboot
[see notes at the bottom if you're feeling adventurous], I don't dare
test it when a reboot's far from likely to hurt me), as long as you
can resist making changes to files in /usr between starting and your
reboot. The difficulty you run into is that you can't copy from /usr/*
(the data on the usr partition) directly into /usr (a folder acting as
a mount point on the root partition). What you *can* do, however, is
bind mount your root partition somewhere else, and your usr partition
too if you like, make the copy (using tar, rsync, or cp with the
appropriate flags to preserve all the important details), adjust
fstab, and then reboot into the adjusted system, after which, you can
go about pondering what to do with your old /usr partition.
Incidentally, this general method also works fairly well for taking
snapshots of a live root partition (accepting that the contents of
anything prone to change mid-copy could be anything, like /tmp, most
of /var, bits of /etc), catching the static files hidden by mounting
done at boot time, like the base set of device nodes in /dev too...

As root, obviously, and you may include --acls or --xattrs  on the
rsync call, if they suit your system's needs (I have quite a bit
turned off on the machine I'm testing this on as I write it up):

1) make backups and... 1.a) verify backups... consider this my disclaimer...
2) mkdir /mnt/root-bind/ /mnt/usr-bind/
3) mount --bind / /mnt/root-bind/
4) mount --bind /usr/ /mnt/usr-bind/
5) rsync --archive --hard-links --sparse --progress /mnt/usr-bind/
/mnt/root-bind/usr/
6) edit fstab ...
7) cross fingers and reboot ...
8) rm -r /mnt/root-bind/ /mnt/usr-bind/

While 9 can be done before the reboot, this way doesn't depend on
fighting with unmounting outside of the reboot itself.

And some background on my test system...
<Before Copy>
# df -h
Filesystem      Size  Used Avail Use% Mounted on
rootfs          7.9G  128M  7.7G   2% /
/dev/root       7.9G  128M  7.7G   2% /
rc-svcdir       1.0M   56K  968K   6% /lib64/rc/init.d
udev             10M  176K  9.9M   2% /dev
shm             2.0G     0  2.0G   0% /dev/shm
/dev/sda1       505M   15M  465M   3% /boot
/dev/sda8        15G  845M   14G   6% /var
/dev/sda7        20G  2.0G   18G  10% /usr
/dev/sda9        49G  3.2G   46G   7% /home
tmpfs           2.0G  4.0K  2.0G   1% /tmp

<After copy (extras trimmed)>
 # df -h
Filesystem      Size  Used Avail Use% Mounted on
rootfs          7.9G  2.1G  5.8G  27% /
/dev/sda7        20G  2.0G   18G  10% /usr

<After reboot>

# df -h
Filesystem      Size  Used Avail Use% Mounted on
rootfs          7.9G  2.0G  5.9G  26% /
/dev/root       7.9G  2.0G  5.9G  26% /
rc-svcdir       1.0M   56K  968K   6% /lib64/rc/init.d
udev             10M  176K  9.9M   2% /dev
shm             2.0G     0  2.0G   0% /dev/shm
/dev/sda1       505M   15M  465M   3% /boot
/dev/sda8        15G  844M   14G   6% /var
/dev/sda9        49G  3.2G   46G   7% /home
/dev/sda10      141G  7.9G  133G   6% /data
tmpfs           2.0G     0  2.0G   0% /tmp

And... at least for my own, copied, rebooted, and working as though
nothing's changed (aside from a presently unused 20G partition sitting
around on the drive). I do admit, my 8G existing root partition is a
bit on the large size.. this system transitioned from a desktop to a
'server', losing X as well as a number of hefty things that were in
/opt ... leaving me with both a lot of free space on / and a much more
slim /usr so it's a rare thing to be able to so easily drop things in
place. It did manage a downtime of all of about a minute, though,
rather than booting a secondary OS to do all of that.

[notes regarding a potentially bad idea follow]
Lastly, on the topic of avoiding the reboot altogether, the trick
would be, instead of unmounting those bind mounts, doing a umount -l
/usr then, as you're able, freeing references to the old mount,
syncing new changes from the old tree into the new one (if any, and if
appropriate), and then starting up whatever was killed to free it, if
appropriate. Due to the usual way /usr is used, however, it's entirely
likely that, after the lazy unmount, the process of simply updating
the system over time will eventually free up all old references,
releasing the old mount, and allowing everything to roll over to the
new arrangement gradually, much like the ability to update things like
sshd in place, restarting the service (as the , but leaving existing
connections uninterrupted... but, again, I'm not brave enough to test
it, let alone make guarantees.

-- 
Poison [BLX]
Joshua M. Murphy

Re: [gentoo-user] Initramfs or move /usr to /, oh my...

[Sebastian Pipping]

On 03/21/2012 03:02 AM, Joshua Murphy wrote:
> 1) make backups and... 1.a) verify backups... consider this my disclaimer...
> 2) mkdir /mnt/root-bind/ /mnt/usr-bind/
> 3) mount --bind / /mnt/root-bind/
> 4) mount --bind /usr/ /mnt/usr-bind/
> 5) rsync --archive --hard-links --sparse --progress /mnt/usr-bind/
> /mnt/root-bind/usr/

At first it looked like you were copying things onto themselves.  Then I
noticed I was wrongly assuming --rbind mount operation rather than the
actual plain --bind that you use.  Nice approach!

Best,



Sebastian

Re: [gentoo-user] Initramfs or move /usr to /, oh my...

[Neil Bothwick]

[-- Attachment #1: Type: text/plain, Size: 802 bytes --]

On Tue, 20 Mar 2012 22:02:31 -0400, Joshua Murphy wrote:

> 1) make backups and... 1.a) verify backups... consider this my
> disclaimer... 2) mkdir /mnt/root-bind/ /mnt/usr-bind/
> 3) mount --bind / /mnt/root-bind/
> 4) mount --bind /usr/ /mnt/usr-bind/

4a) mount /usr -o remount,ro

Makes sure nothing changes on /usr between starting the sync and
rebooting.

> 5) rsync --archive --hard-links --sparse --progress /mnt/usr-bind/
> /mnt/root-bind/usr/
> 6) edit fstab ...
> 7) cross fingers and reboot ...
> 8) rm -r /mnt/root-bind/ /mnt/usr-bind/

Create the temporary mounts points in /tmp and you save this step - since
I added one, it is only fair to remove one to not increase the workload :)


-- 
Neil Bothwick

When companies ship Styrofoam, what do they pack it in?

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [gentoo-user] Re: Initramfs or move /usr to /, oh my...

[J. Roeleveld]

On Mon, March 19, 2012 1:31 am, Pandu Poluan wrote:
> When I started administering remote servers, Citrix's XenServer is Good
> Enoughâ„¢ to deploy in production, so now it's the first thing I install
> on a
> virgin box, even if said virgin box will host only one VM.
>
> This provides me with a usable Virtual Console through which I can watch
> the boot process.

These things are soo usefull :)
That's one of the reasons why I have decided to only get servers with
remote-desktop-over-network support in hardware :)

I don't want to leave a screen and keyboard connected to machines that are
supposed to run independently.

--
Joost