bug#65378: time-machine v1.0.0

[Simon Tournier]

Hi,

Thanks for the report.

On Sat, 19 Aug 2023 at 10:59, Théo Maxime Tyburn <theo.tyburn@gmail.com> wrote:

> 4 tests failed:
>     test_ftplib test_httplib test_ssl test_urllib2_localnet

[...]

> srfi/srfi-1.scm:592:17: In procedure map1:
> Throw to key `srfi-34' with args `(#<condition &store-protocol-error 
> [message: "build of 
> `/gnu/store/sp947xhp8dqfznzd2m0aq4ia5qvklbl-python2-2.7.15.drv' failed" 
> status: 100] 7f5cc6830cf0>)'.
> guix time-machine: error: You found a bug: the program 
> '/gnu/store/d12x45lz2dv3mgp594kzjv0d121g0ncs-compute-guix-deivation'
> failed to compute the derivation for Guix (version: 
> "6298c3ffd9654d3231a6f25390b056483e8f407c"; system: "x86_64-linx";
> host version: "4547bc6fa3142dca77f7fc912368aeff31bd6e53"; pull-version: 1).

That’s a known bug.  It’s a time-bomb in the test suite of some packages
(here Python 2).  Other said, the bug is from upstream. ;-)

This failure cannot be fixed and the only way is to set the hardware
clock to some past date (pick 2019 or 2020).

Well, one thing we could do is to set a future date when building with
the CI in order to detect potential problems.  It also would help with
the 2038 bug [1]. ;-)

>From my point of view, I do not know if it is worth to keep this bug
open since it will not be fixed.  And, it is documented in the manual,
see the second note of [2].

What do you think about closing?


1: https://en.wikipedia.org/wiki/Year_2038_problem
2: https://guix.gnu.org/manual/devel/en/guix.html#Invoking-guix-time_002dmachine

Cheers,
simon