all messages for Guix-related lists mirrored at yhetil.org
 help / color / mirror / code / Atom feed
From: Julien Lepiller <julien@lepiller.eu>
To: help-guix@gnu.org,Vladilen Kozin <vladilen.kozin@gmail.com>
Subject: Re: How to debug failed boot after successful reconfigure?
Date: Fri, 09 Apr 2021 15:36:59 -0400	[thread overview]
Message-ID: <423006D4-2E90-4FAA-BB4C-F834BAE33A96@lepiller.eu> (raw)
In-Reply-To: <CACw=CXPZVFSA-Ha6hYfnR58-+FoPQpHTXH0Rv49oC8MGgWmxNA@mail.gmail.com>

Hi, help-guix is the right mailing, but we're not always very responsive, plus there is a delay for your first mail to pass spam filter.

In /var/guix/profiles/system-<n>-link (where n corresponds to the generation you wart to inspect), you will hind an etc directory.

dmesg accumulates in /var/log/messages, I don't think there's a way to find it for a given generation, but you should find the relevant lines by looking at the time they were registered.

HTH!

Le 9 avril 2021 07:53:16 GMT-04:00, Vladilen Kozin <vladilen.kozin@gmail.com> a écrit :
>Hm, bit puzzled. Noone ever encountered boot failures with Guix?
>Should I've sent it to guix-devel?
>
>I'm going to try my luck in a separate thread with a much narrower
>question.
>
>FWIW for the issue at hand I ended up leveraging the fact that I'm
>running on a server with out-of-band ethernet and IPMI on board.
>Essentially, I ended up configuring that server to communicate Serial
>over Lan, then reconfigured Guix with Grub and Kernel both doing IO to
>serial console, then connected over SSH while running inside a Tmux
>session ... so now I can scroll up if boot fails and see exactly what
>happened. This would probably only work on an actual enterprise hw
>that lets you do IPMI SOL. For anyone interested here're relevant
>parts of my config.scm:
>
>;; make sure kernel IO to serial both com1 and com2
>(kernel-arguments '("console=tty1 console=ttyS0,115200n8
>console=ttyS1,115200n8"))
>
>;; make sure grub IO goes over serial both com1 and com2
>(bootloader
>  (bootloader-configuration
>   (bootloader grub-bootloader)
>   ;; TODO well that's nuts given that plugging in USB drive changes
>   ;; order of block devices
>   (target "/dev/sda")
>   (keyboard-layout keyboard-layout)
>   (terminal-inputs '(console serial_0 serial_1))
>   (terminal-outputs '(console serial_0 serial_1))))
>
>Seriously, noone knows where to find dmesg and contents of /etc that
>pertain to the last attempt to boot specific generation?
>Pretty please?
>
>On Thu, 8 Apr 2021 at 16:58, Vladilen Kozin <vladilen.kozin@gmail.com>
>wrote:
>>
>> Hi list.
>>
>> I'm trying to wrap my head around running guix. While question in
>subj
>> is general it was prompted by a real failure, so really I am
>> interested both in "general approach" as well as solving the actual
>> problem I ran into.
>>
>> having updated with `guix pull` I run `sudo guix system reconfigure
>> config.scm` which succeeds. I reboot and the boot fails attempting to
>> mount the new file-systems I've introduced in that config. This being
>> guix I reboot into older generation. However, now I am unsure how to
>> debug what'd happened. E.g. I would like to:
>> - look at failed dmesg or equivalent log,
>> - look at files that reconfigure produced in /etc, e.g. /etc/fstab
>>
>> IIUC when I boot even from an older generation, the failed one will
>be
>> marked as "current" (that is the newest reconfigured) but everything
>> in /etc appears to belong to the older generation - hardly
>surprising.
>> Following link shown by `guix system describe` in my case
>> /var/guix/profiles/system-2-link I don't see e.g. /etc/fstab I'm
>> interested in. Where are those files?
>>
>> So, the questions above (maybe others I should've asked?) amount to
>my
>> asking, how to debug the failure.
>>
>> My specific case was adding these files-systems to my os declaration:
>> (file-system
>>     (mount-point "/mnt/tempb")
>>     (device (file-system-label "tempb"))
>>     (type "ext4")
>>     #;(flags '(no-atime))
>>     (options "defaults,noatime,discard,user")
>>     (create-mount-point? #t))
>>
>> At boot time I got multiple errors from mount unable to figure the
>> options supplied. This in itself is puzzling seeing how I did check
>> /etc/fstab after reconfigure and was able to manually mount as per
>> that generated /etc/fstab without issue. One note is that I think the
>> (create-mount-point? #t) had not been honored, but I did create those
>> mount points manually. Dunno, might be a bug in guix proper. But
>> that's minor.
>>
>> Thanks
>>
>> --
>> Best regards
>> Vlad Kozin
>
>
>
>-- 
>Best regards
>Vlad Kozin

      reply	other threads:[~2021-04-09 19:39 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-04-08 15:58 How to debug failed boot after successful reconfigure? Vladilen Kozin
2021-04-09 11:53 ` Vladilen Kozin
2021-04-09 19:36   ` Julien Lepiller [this message]

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

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

  git send-email \
    --in-reply-to=423006D4-2E90-4FAA-BB4C-F834BAE33A96@lepiller.eu \
    --to=julien@lepiller.eu \
    --cc=help-guix@gnu.org \
    --cc=vladilen.kozin@gmail.com \
    /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 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.