unofficial mirror of guix-devel@gnu.org 
 help / color / mirror / code / Atom feed
From: Julien Lepiller <julien@lepiller.eu>
To: Dan Partelly <dan_partelly@rdsor.ro>
Cc: guix-devel@gnu.org
Subject: Re: Suboptimal experience
Date: Mon, 18 Jun 2018 17:52:10 +0200	[thread overview]
Message-ID: <d26208b96c61eb788214074d106bdf4d@lepiller.eu> (raw)
In-Reply-To: <1202EA49-0D08-4EAB-A128-F9C7FBBA0EEC@rdsor.ro>

Le 2018-06-18 17:01, Dan Partelly a écrit :
> Well, retried again a git pull today. It broke with no backtrace (an
> exception while printing backtrace)
> ERROR: In procedure private-lookup: No variable bound to
> define-module* in module (guile)
> 
> cannot build derivation for /gnu/store/sha…guix-32eb44240-modules.drv:
> 1 dependencies could not be built
> cannot build derivation for /gnu/store/sha…guix-32eb44240.drv: 1
> dependencies could not be built

This is a bug in guix pull. It has already been reported and we're 
working
on a fix. That shouldn't cause any issue on your side: the failure to 
pull
simply means that your guix wasn't updated. As a workaround, you can try 
again
as long as it fails in this way. "guix pull --cores=1" might help too.

The error is not in the package definitions, it's more likely a race 
condition
introduced with the recent guix pull update.

> 
> I do not know whats going on, but one two things are sure:
> 
> 1. If is true that any packager can break the build of guix by
> commiting a bad package definition, then the whole packaging system
> design is **FLAWED** in my opinion. A system which is so brittle has
> no place in an industrial environment, at least until such issues are
> permanently resolved.  Yes, Mark was nice enough to say you can build
> it from source and it works better, but to be honest, when I work with
> an OS I just want tools which do work and make my life easier, not
> spit error after error for days. If I need to build a repository, it
> must be because i want to modify something or customize some software,
> not because the OS itself burps .

Sorry about the bad experience :/ As I said, all this is due to the fact 
that
guix evolves really fast. That's also the reason why we still consider 
it
beta. Had you tested it a few days earlier, everything would have been
going smoothly.

> 
> 2. A package manager which fails time and again to build itself
> doesn't present itself as very trustworthy. It may be very well be,
> but adoption of software is governed by social factors more than
> technical excellence. GuixSD for now is very nice and cool, but its a
> hacking playground, not an industrial strength OS which you deliver
> without worry.
> 
> So please excuse me if I give a bit of unsolicited advice, please fix
> all those before going 1.0.  It does not gives a good impression on
> the OS. I believe its important for your future marketing and your
> goals to further free software in this world.

Thank you for your advice :) I think you're right: we are still beta, so
errors are to be expected from time to time. We should fix them before 
we
go 1.0. That said, the system is robust enough that you can always 
recover
from said errors.

> 
> 
> 
>> On Jun 18, 2018, at 14:11, swedebugia <swedebugia@riseup.net> wrote:
>> 
>> Hi
>> 
>> On June 17, 2018 9:51:51 PM GMT+02:00, Dan Partelly 
>> <dan_partelly@rdsor.ro> wrote:
>>> HI Mark
>> 
>> [...]
>> 
>>> For any “guix system …. “ the message was:
>>> 
>>> guix: system: command not found .
>> 
>> I experienced this as well today after guix pull from 0.14 to commit 
>> http://git.savannah.gnu.org/cgit/guix.git/commit/?id=b4eae997fe5b928f179c34d281e9f2c3eccd3670 
>> if I remember correctly.
>> 
>> I just understood from the previous reply that running guix pull again 
>> will solve the problem. I will try later and report back the result.
>> 
>> --
>> Cheers Swedebugia

  reply	other threads:[~2018-06-18 15:52 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-06-17 10:09 Suboptimal experience Dan Partelly
2018-06-17 15:40 ` Julien Lepiller
2018-06-17 19:55   ` Dan Partelly
2018-06-17 20:09     ` Julien Lepiller
2018-06-17 20:13       ` Dan Partelly
2018-06-17 20:16         ` Julien Lepiller
2018-06-17 17:31 ` Mark H Weaver
2018-06-17 19:51   ` Dan Partelly
2018-06-18 11:11     ` swedebugia
2018-06-18 15:01       ` Dan Partelly
2018-06-18 15:52         ` Julien Lepiller [this message]
2018-06-18 18:08           ` Dan Partelly
2018-06-18 18:29             ` Nils Gillmann

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: https://guix.gnu.org/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=d26208b96c61eb788214074d106bdf4d@lepiller.eu \
    --to=julien@lepiller.eu \
    --cc=dan_partelly@rdsor.ro \
    --cc=guix-devel@gnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
Code repositories for project(s) associated with this public inbox

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

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).