bug#27976: install 32b xfce no gnome: collision hicolor/icon-theme.cache

bug#27976: install 32b xfce no gnome: collision hicolor/icon-theme.cache

[Paul Dufresne]

[-- Attachment #1: Type: text/plain, Size: 758 bytes --]

After having problem with adwaita-icon-them-3.24.0, (reported in other bug)
 I decided to give a try by removing gnome service from the
list of "meta-package" to install.

So in fact keeping 'only' XFCE.

I tried two times, once after the adwaita problem, the other after a reboot
and from scratch (except partitioning).

The result was the same (from poorly taken notes):
...
warning: collision encountered:
/gnu/store/7qnfb279...hicolor/icon-theme.cache
...
warning: arbitrarily choosing 7qnfb...icons/hicolor/icons-theme.cache
...
Backtrace:
In Ice-0/boot-9.scm
...
In unknown file
   gtk-in-modules-builder

That's it, what I had noted.

So, I had to boot with an other system to report this, was unable to get
guix-sd 0.13.0 usb install 32b to install.

[-- Attachment #2: Type: text/html, Size: 1019 bytes --]

bug#27976: install 32b xfce no gnome: collision hicolor/icon-theme.cache

[Paul Dufresne]

I installed with 64b.

The collision happened in 64b, but did not stop installation.

bug#27976: install 32b xfce no gnome: collision hicolor/icon-theme.cache

[Ludovic Courtès]

Paul Dufresne <dufresnep@gmail.com> skribis:

> After having problem with adwaita-icon-them-3.24.0, (reported in other bug)
>  I decided to give a try by removing gnome service from the
> list of "meta-package" to install.
>
> So in fact keeping 'only' XFCE.
>
> I tried two times, once after the adwaita problem, the other after a reboot
> and from scratch (except partitioning).
>
> The result was the same (from poorly taken notes):
> ...
> warning: collision encountered:
> /gnu/store/7qnfb279...hicolor/icon-theme.cache
> ...
> warning: arbitrarily choosing 7qnfb...icons/hicolor/icons-theme.cache

This is a harmless warning.

> Backtrace:
> In Ice-0/boot-9.scm
> ...
> In unknown file
>    gtk-in-modules-builder

This looks like the real error, but unfortunately we don’t have enough
details here to see what the failure was.  It would have been helpful to
see the whole backtrace or the actual “ERROR” line, along with the line
that says precisely which derivation failed (though I understand it’s
hard to grab that info while you’re installing.)

Thanks,
Ludo’.