[Libre-soc-dev] Coriolis2 Non-Chroot Script

Andrey Miroshnikov andrey at technepisteme.xyz
Sat Sep 11 21:52:24 BST 2021


On 11/09/2021 19:29, Andrey Miroshnikov wrote:
 > Just giving "mk-deb-chroot" a read, I noticed lines 25-30, 57-62, 69-75,
 > 79-88, would crash the script (as they are not valid commands, or
 > continued by backslashes, or encapsulated by quotes).
Replying to myself after I realised I was being too much of a newbie. 
Just tested the instruction using concatenate:
cat <<EOF >> file_to_fill
And realised that whatever occurs afterwards will be recorded into 
file_to_fill until EOF is entered. Fascinating.


On 11/09/2021 19:52, lkcl wrote:
> it's only critically important for the creation of the GDS-II files
Ok, then I'll use chroot going forwards.

> 
> but, also, i have found that in some instances software won't even compile on debian/testing, and conversely some won't compile on debian/stable.

Ah, the joys of stable/unstable repos. One of the reasons Debian is one 
of the last distros I started using was because I found it more 
difficult to setup software I needed as a general user.

Apt just seems to be far less intuitive than some of the other package 
managers (although it's probably due to my inexperience).
However you can't argue with sheer number of packages and support/mirror 
availability of Debian.


> the other reason is, some people use ubuntu and doing so results in the wrong versions of libraries. a chroot is "stable common ground" that allows continuing to work.
> 
> thus it is down to need and circumstances.
Makes sense. Environment consistency will lead to more sanity (hopefully).


I'll get "mk-deb-chroot" running first, then look into coriolis2.

Does "mk-deb-chroot" still needed the "/tmp/brokenproxy" section? Will 
apt always have issues with https?


Andrey



More information about the Libre-soc-dev mailing list