Summary: | glsa 201701-75: Wrong instruction on how to update dev-lang/perl. | ||
---|---|---|---|
Product: | Gentoo Security | Reporter: | Teika kazura <teika> |
Component: | GLSA Errors | Assignee: | Gentoo Security <security> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | kentnl |
Priority: | Normal | ||
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- |
Description
Teika kazura
2017-04-01 05:35:10 UTC
I'm not sure it qualifies as "Incorrect". The problem is portage does the wrong thing, even with the "most correct" invocation there is. Subsequently, there is no single "correct" approach, just a collection of different approaches for different problems. For instance, recently, it took a full day of supporting a user in #gentoo in order to coerce their system into doing the right thing. I know this isn't great, but we're ultimately limited by what portage lets us do, and the complexities involved with user-choice makes even the most ideal instructions fail. Maybe I was ambigous. Ideal solutions have to be sought, but it's, in practice, unfeasible right now. I don't know how to set up the goal properly. Yet, something practical is necessary, so that users won't be left at loss. Actually I was. (That's what caused me to create the wiki page "Perl") How about mentioning that Wiki page? You (kentnl) and Dilfrige have seen and revised it, and it describes several ways to cope with subslot conflict. Though Wiki has no guarantee of the contents stability, a specific revision can be linked to. I know GLSA is official, and it's better to avoid hacks, and should be definitive. However the solution *has to be* effective. There must be some good idea. ;) Thanks for the report. I updated the GLSA to address your concerns: https://gitweb.gentoo.org/data/glsa.git/commit/?id=c0ee40ad4dc5c96ebf25659e3374fc38f9a5a4a2 |