[gentoo-amd64] Re: amd64 list, still useful? Was: btrfs

[Duncan]

Mark Knecht posted on Wed, 04 Jun 2014 09:41:30 -0700 as excerpted:

> There is an in progress, higher energy thread on gentoo-user with folks
> getting upset (my interpretation) about systemd and support for
> suspend/resume features. I only found it being that I ran into an emerge
> block and went looking for a solution. (In my case it was -upower as a
> new use flag setting.)

Yeah.  I saw the original dev-list thread on the topic, before it all hit 
the tree (and continuing now), which is a big part of why I subscribe to 
the dev-list, to get heads-up about things like that.

What happened from the dev-list perspective is that after upower dropped 
about half the original package as systemd replaced that functionality, 
the gentoo maintainers split the package in half, the still included 
functionality under the original upower name, with the dropped portion in 
a new, basically-gentoo-as-upstream, package, upower-pm-utils.

But to the gentoo maintainer the portage output was sufficient that 
between emerge --pretend --tree --unordered-display and eix upower, what 
was needed was self-evident, so he didn't judge a news item necessary.  
What a lot of other users (including me) AND devs are telling him is that 
he's apparently too close to the problem to see that it's not as obvious 
as he thinks, and a news item really is necessary.

Compounding the problem for users is that few users actually pulled in 
upower on their own and don't really know or care about it -- it's pulled 
in due to default desktop-profile use-flags as it's the way most desktops 
handle suspend/hibernate.  Further, certain desktop dependencies 
apparently got default-order reversed on the alternative-deps, so portage 
tries to fill the dep with systemd instead of the other package.  
Unfortunately that's turning everybody's world upside down, as suddenly 
portage wants to pull in systemd *AND* there's all these blockers!

Meanwhile, even tho he didn't originally think it necessary, once pretty 
much all gentoo userspace (forums, irc, lists, various blogs...) erupted 
in chaos, the gentoo maintainer decided that even tho he didn't quite 
understand /why/ a news item was needed, that was the best way to get the 
message out as to how to fix things and to calm things back down.

But, policy is that such news items must be posted to the gentoo-dev list 
for (ideally) several days of comment before they're committed, and a 
good policy it is in general too, because the news items generally turn 
out FAR better with multiple people looking over the drafts and making 
suggestions, than the single-person first-drafts tend to be!

In cases such as this, however, the comment time is shortened to only a 
day or two unless something seriously wrong comes up in the process, and 
while I've not synced for a few days, I'd guess that news item has either 
hit before I send this, or certainly if not, it'll hit within a matter of 
hours.

Once the news item hits, for people that actually read them at least, the 
problem should be pretty much eliminated, as there's appropriate 
instructions for how to fix the blocker, etc.

So things should really be simmering back down pretty shortly. =:^)  
Meanwhile, in the larger perspective of things, it's just a relatively 
minor goof that as usual is fixed in a couple days.  No big deal, except 
that /this/ goof happens to include the uber-lightening-rod-package that 
is systemd.  Be that as it may, the world isn't ending, and the problem 
is indeed still fixed up within a couple days, as usual, with 
information, some reliable, some not so reliable, available via the usual 
channels for those who don't want to wait.

-- 
Duncan - List replies preferred.   No HTML msgs.
"Every nonfree program has a lord, a master --
and if you use the program, he is your master."  Richard Stallman