--- Begin Message ---
Subject: |
User account password got locked when booting old generation |
Date: |
Wed, 29 May 2019 22:45:17 +0200 |
User-agent: |
NeoMutt/20180716 |
After I reconfigured to a broken, unbootable generation and then
rebooted to an old working generation, I found my user account
password was locked. This was my /etc/shadow:
root::18045::::::
florian:!:18045::::::
nobody:!:18045::::::
guixbuilder01:!:18045::::::
guixbuilder02:!:18045::::::
guixbuilder03:!:18045::::::
guixbuilder04:!:18045::::::
guixbuilder05:!:18045::::::
guixbuilder06:!:18045::::::
guixbuilder07:!:18045::::::
guixbuilder08:!:18045::::::
guixbuilder09:!:18045::::::
guixbuilder10:!:18045::::::
ntpd:!:18045::::::
messagebus:!:18045::::::
polkitd:!:18045::::::
geoclue:!:18045::::::
colord:!:18045::::::
avahi:!:18045::::::
gdm:!:18045::::::
httpd:!:18045::::::
Logging in as root (root had an empty password before as well) and
running `passwd florian` fixed it, and I *cannot* reproduce the bug
anymore by booting the broken generation again, i.e. my password
remains set now.
I presume it is not possible to lock a password by typing the wrong
password too often on the virtual console?
If you think this is not enough material to work on, feel free to
close this bug, but there seems to be some misbehavior somewhere in
Guix’ password management.
(The reason for the new generation’s brokenness seems unrelated; it
could not boot after I tried adding syslogd to the requirements of
udev-shepherd-service.)
Regards,
Florian
--- End Message ---
--- Begin Message ---
Subject: |
Re: bug#35996: User account password got locked when booting old generation |
Date: |
Thu, 06 Jun 2019 10:04:33 +0200 |
User-agent: |
Gnus/5.13 (Gnus v5.13) Emacs/26.2 (gnu/linux) |
Hello!
"pelzflorian (Florian Pelz)" <address@hidden> skribis:
> On Wed, Jun 05, 2019 at 11:13:34PM +0200, Ludovic Courtès wrote:
[...]
>> The attached patch for Shepherd moves everything before loading the
>> config file. I think it will have the desired effect, though I’m not
>> entirely sure the signal handler would run at the right time etc.
>>
>
> It works for me without recovering journal message (and taking an
> insignificantly longer time to reboot).
Excellent. Pushed as Shepherd commit
c6f250d1fd1afa9ee49c8bb2414eee087b672789.
Thank you!
Ludo’.
--- End Message ---