all messages for Guix-related lists mirrored at yhetil.org
 help / color / mirror / code / Atom feed
* 'guix system vm' questions
@ 2016-06-29  6:57 myglc2
  2016-06-29 14:10 ` Thompson, David
  0 siblings, 1 reply; 14+ messages in thread
From: myglc2 @ 2016-06-29  6:57 UTC (permalink / raw)
  To: help-guix

[-- Attachment #1: Type: text/plain, Size: 1001 bytes --]

I have a headless server running Guix/Debian 8.3 on which I would like
to run some guix vms. So far, I have a script (see mkvm.sh, attached)
that runs a single vm. It has some issues:

1) lsh-service hangs waiting for keystrokes on the QEMU console

2) passwords need to be set in the QEMU console

3) on the vm, 'guix package -i' throws an error, e.g.,

   (on the console)
   v1 login: u1
   password: <password>
   $ guix package -i
   accepted connection from pid 390, user u1
   guix package: error: build failed: changing ownership of '/path/store': Read-only file system

4) I can't log in to the vm like I expected, e.g. 'ssh u1@10.0.2.15'
   times out

5) I would like to have the user 'u1' share read/write access to the
   $HOME directory and be able to use $HOME/.ssh credentials

Your suggestions &/or sample configurations would be most welcome.

TIA - George

PS I am running this guix git checkout ...
   * master                                  d6e374a gnu: mafft: Update to 7.299.


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: mkvm.sh --]
[-- Type: text/x-sh, Size: 1484 bytes --]

#!/bin/bash

# make and run a Guix VM

set -x
VMNUM=1
tmpdir=vm$VMNUM
rm -fr $tmpdir
mkdir $tmpdir

# vm.config
cat > "$tmpdir/vm.scm"<<EOF
(use-modules (gnu))
(use-service-modules networking ssh)
(use-package-modules admin
		     disk
		     ssh rsync wget screen
		     version-control
		     emacs
		     xorg certs
		     graphviz
		     package-management
		     )
(operating-system
  (host-name "v1")
  (timezone "America/New_York")
  (locale "en_US.utf8")
  (bootloader (grub-configuration (device "/dev/sda")))
  (file-systems (cons (file-system
			(device "g1sd")
			(title 'label)
			(mount-point "/")
			(type "ext4"))
		      %base-file-systems))
  (users (cons* (user-account
		 (name "u1")
		 (comment "u 1")
		 (group "users")
		 (supplementary-groups '("wheel"))
		 (home-directory "/home/u1"))
		%base-user-accounts))
  (packages
   (cons*
    glibc-utf8-locales
    parted
    openssh nss-certs xauth rsync wget git git-manpages
    emacs magit screen 
    graphviz
    guix
    %base-packages
    )
   )
  (services (cons* (dhcp-client-service)
		   (lsh-service #:root-login? #t)
		   %base-services)))
EOF

# show guix version
# guix system --version
(stat $HOME/.config/guix/latest | grep File:)
git -C ~/.config/guix/latest branch -av | grep '* master'
make -C $HOME/.config/guix/latest -s

# make vm & run
# Note: '-k en-us' is a hack to get Mac XQuartz keys to sort of work
$(guix system vm $tmpdir/vm.scm --expose=$HOME --share=$HOME/tmp=/exchange) -k en-us

^ permalink raw reply	[flat|nested] 14+ messages in thread
* Re: 'guix system vm' questions
@ 2016-06-29 18:23 David Craven
  2016-06-30 13:10 ` David Craven
  0 siblings, 1 reply; 14+ messages in thread
From: David Craven @ 2016-06-29 18:23 UTC (permalink / raw)
  To: help-guix

HW_RANDOM_VIRTIO is enabled in the default kernel.

Starting the vm like this enables the hwrng:
/gnu/store/3rqbwx34kfa789jbywfbxl90nids08cf-run-vm.sh -device virtio-rng-pci
and it works as can be verified by
cat /dev/hwrng

The only missing components from what I can tell is packaging rng-tools
and adding a service to add /dev/hwrng to the entropy pool on boot.

https://git.kernel.org/cgit/utils/kernel/rng-tools/rng-tools.git/

> We should look into QEMU's VirtIO RNG, which could help here:
> http://wiki.qemu-project.org/Features-Done/VirtIORNG

^ permalink raw reply	[flat|nested] 14+ messages in thread
* Re: 'guix system vm' questions
@ 2016-07-03 16:09 David Craven
  2016-07-06 16:17 ` myglc2
  0 siblings, 1 reply; 14+ messages in thread
From: David Craven @ 2016-07-03 16:09 UTC (permalink / raw)
  To: help-guix

@myglc2 the simplest workaround to this issue is to add -device
virtio-rng-pci to your qemu flags and patch lsh to use /dev/hwrng in
the /bin/lsh-make-seed file I linked to in a previous reply.

On Sun, Jul 3, 2016 at 6:00 PM,  <help-guix-request@gnu.org> wrote:
> Send Help-Guix mailing list submissions to
>         help-guix@gnu.org
>
> To subscribe or unsubscribe via the World Wide Web, visit
>         https://lists.gnu.org/mailman/listinfo/help-guix
> or, via email, send a message with subject or body 'help' to
>         help-guix-request@gnu.org
>
> You can reach the person managing the list at
>         help-guix-owner@gnu.org
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Help-Guix digest..."
>
>
> Today's Topics:
>
>    1. Re: 'guix system vm' questions (myglc2)
>    2. Re: 'guix system vm' questions (myglc2)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Sun, 03 Jul 2016 09:47:42 -0400
> From: myglc2 <myglc2@gmail.com>
> To: help-guix@gnu.org
> Subject: Re: 'guix system vm' questions
> Message-ID: <86mvly6dfl.fsf@gmail.com>
> Content-Type: text/plain; charset=utf-8
>
> ludo@gnu.org (Ludovic Court?s) writes:
>
>> myglc2 <myglc2@gmail.com> skribis:
>>
>>> Leo Famulari <leo@famulari.name> writes:
>>>
>>>> On Wed, Jun 29, 2016 at 10:10:35AM -0400, Thompson, David wrote:
>>>>> On Wed, Jun 29, 2016 at 2:57 AM, myglc2 <myglc2@gmail.com> wrote:
>>>>> > I have a headless server running Guix/Debian 8.3 on which I would like
>>>>> > to run some guix vms. So far, I have a script (see mkvm.sh, attached)
>>>>> > that runs a single vm. It has some issues:
>>>>> >
>>>>> > 1) lsh-service hangs waiting for keystrokes on the QEMU console
>>>>>
>>>>> LSH needs to be initialized with a key on first boot, which is why you
>>>>> need to type to create entropy.  This sucks for automation, so I would
>>>>> recommend OpenSSH instead, but we don't have an openssh-service yet.
>>>>
>>>> In your operating system configuration, you can pass 'initialize? #f' to
>>>> lsh-service to skip the SSH seed and host-key initialization. You will
>>>> need to initialize the LSH service later, before you can use it.
>>>
>>> I tried that and ... lsh spun looking for a seed file advising me to run
>>> 'lsh-make-seed', and then stopped.
>>
>> Yeah this is expected: lshd won?t start if it cannot find a seed;
>> eventually, the Shepherd?
>>
>>> Service SSH-daemeon has started.
>>> No seed file. Please create on by running
>>> lwh-make-seed -o "/var/spool/lsh/yarrow-seed-file".
>>> lshd: No rendomness generator available.
>>> Service ssh-daemon has been disabled.
>>>   (Respawning too fast.)
>>
>> ? disables it.
>>
>>> v1 login: root
>>> root@v1 ~# lsh-make-seed
>>> -bash: lsh-make-seed: command not found
>>
>> I guess ?lsh-service-type? should be changed to extend
>> ?profile-service-type? such that this command is available.  WDYT?
>
> Well I can't see how it will help in my use case. I can't think of why
> it would be useful in practice, since lsh-make-seed is already being
> called with --sloppy by 'ssh-service initialize? #t'.
>
> The only reasons I can think of are ...
>
> - to enable a user at the console to manually reproduce what
>    'ssh-service initialize? #t' does, and
>
> - to avoid it being reported as a bug
>
> ... which is why I didn't report it as a bug.
>
> Re my use case, am I correct in believing that the only way to address
> it is to enhance lsh-service with an alternative initialization scheme
> that does not require console input?
>
> If so, maybe we should rename the current initialize to ...
>
> INITIALIZE-REQUIRING-CONSOLE-INPUT?
>
> ... and add a new one named ...
>
> INITIALIZE-UNATTENDED?
>
> WDYT? - George
>
>
>
>
>
>
>
>
>
> ------------------------------
>
> Message: 2
> Date: Sun, 03 Jul 2016 09:53:28 -0400
> From: myglc2 <myglc2@gmail.com>
> To: help-guix@gnu.org
> Subject: Re: 'guix system vm' questions
> Message-ID: <86inwm6d5z.fsf@gmail.com>
> Content-Type: text/plain
>
> David Craven <david@craven.ch> writes:
>
>> So I packaged rng-tools and created a shepherd service, but I realized
>> that the service needs to be started before the activation part of the
>> lsh service. What's the best way to do this?
>>
>> https://git.lysator.liu.se/lsh/lsh/blob/master/src/lsh-make-seed.c#L1075
>>
>> On Wed, Jun 29, 2016 at 8:23 PM, David Craven <david@craven.ch> wrote:
>>> HW_RANDOM_VIRTIO is enabled in the default kernel.
>>>
>>> Starting the vm like this enables the hwrng:
>>> /gnu/store/3rqbwx34kfa789jbywfbxl90nids08cf-run-vm.sh -device virtio-rng-pci
>>> and it works as can be verified by
>>> cat /dev/hwrng
>>>
>>> The only missing components from what I can tell is packaging rng-tools
>>> and adding a service to add /dev/hwrng to the entropy pool on boot.
>>>
>>> https://git.kernel.org/cgit/utils/kernel/rng-tools/rng-tools.git/
>>>
>>>> We should look into QEMU's VirtIO RNG, which could help here:
>>>> http://wiki.qemu-project.org/Features-Done/VirtIORNG
>
> This sounds great. I see you submitted patches. Am I correct in thinking
> that the next step is to add an lsh-option option that uses rmg-tools
> instead of lsh-make-seed to set the yarrow seed?
>
> TIA - George
>
>
>
>
> ------------------------------
>
> Subject: Digest Footer
>
> _______________________________________________
> Help-Guix mailing list
> Help-Guix@gnu.org
> https://lists.gnu.org/mailman/listinfo/help-guix
>
>
> ------------------------------
>
> End of Help-Guix Digest, Vol 8, Issue 4
> ***************************************

^ permalink raw reply	[flat|nested] 14+ messages in thread
* Re: 'guix system vm' questions
@ 2016-07-07 16:26 David Craven
  0 siblings, 0 replies; 14+ messages in thread
From: David Craven @ 2016-07-07 16:26 UTC (permalink / raw)
  To: help-guix

> ... I don't follow this. Are you referring to the posts above?

The idea was that replacing /dev/random with /dev/hwrng would do the trick,

https://git.lysator.liu.se/lsh/lsh/blob/master/src/lsh-make-seed.c#L237

but I haven't actually tested it.

I didn't get lsh to work with rngd-service because of the rngd-service
would have to load before the lsh-activation-service, which is not the
case. I there isn't enough entropy in /dev/random it starts it's
lsh-make-seed script which doesn't query /dev/random anymore. This
means that entropy added too late is like no entropy added at all.

There is dropbear-service waiting for review, which works without
rngd-service. If you find test the patch from guix-devel you have to
replace the -net user flag in vm.scm with -net
user,hostfwd=tcp::10022-:22. Just adding it after didn't work. I'll
submit a patch soon to fix that, by using environment variables, so
you can just source qemu-flags.sh.

Cheers
David

^ permalink raw reply	[flat|nested] 14+ messages in thread

end of thread, other threads:[~2016-07-11 19:33 UTC | newest]

Thread overview: 14+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2016-06-29  6:57 'guix system vm' questions myglc2
2016-06-29 14:10 ` Thompson, David
2016-06-29 15:48   ` Leo Famulari
2016-07-01 20:15     ` myglc2
2016-07-02 13:06       ` Ludovic Courtès
2016-07-03 13:47         ` myglc2
2016-07-11 19:32     ` Leo Famulari
2016-06-30 21:20   ` myglc2
  -- strict thread matches above, loose matches on Subject: below --
2016-06-29 18:23 David Craven
2016-06-30 13:10 ` David Craven
2016-07-03 13:53   ` myglc2
2016-07-03 16:09 David Craven
2016-07-06 16:17 ` myglc2
2016-07-07 16:26 David Craven

Code repositories for project(s) associated with this external index

	https://git.savannah.gnu.org/cgit/guix.git

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.