Gentoo Archives: gentoo-user

From: James <wireless@×××××××××××.com>
To: gentoo-user@l.g.o
Subject: [gentoo-user] Re: Project:Installer
Date: Sun, 26 Jul 2015 17:20:16
Message-Id: loom.20150726T191838-154@post.gmane.org
In Reply to: Re: [gentoo-user] Re: Project:Installer by Bruce Schultz
1 Bruce Schultz <brulzki <at> gmail.com> writes:
2
3 >
4 >> Matthew Marchese <maffblaster@g.o> writes:
5
6 >> I see that you've found stager. I'd like you to share
7 your thoughts >> on what a perfect installer Gentoo could do.
8
9 A successful gentoo installer will:
10
11 Be multi-faceted so that many different, but common
12 installation outcomes are not only possible, but are
13 automated to the point of extreme convenience for folks to
14 use them, as they choose. Let's face it no matter what we do,
15 most noobs will not use Gentoo. But, those folks with some
16 level of experience and competence will use gentoo; many more
17 if there is an automated (base)installation. After all, when
18 google or others corporations install and use gentoo, do you
19 think they have folks spend 1-2 days using the handbook? NO,
20 their gentoo(derivative) has an automated installation.
21
22
23 So a base-installer for your [category 1] is the
24 most important part. So in that train of thought,
25 WE, should parse out all of the good parts of many
26 different installers and installation schemes, as a part
27 of the research and leverage as to what exists that can
28 be leveraged or emulated, Debian included. OpenSuse has
29 (13.2) has a slick install that allows for btrfs without
30 lvm or mdadm. That was the default pathway. I've read that
31 you can end up with a full raid install if you choose the
32 "advanced" pathway. I'm still researching that one. Then
33 there is 'Calculate Linux' that more than one gentoo dev
34 uses routinely to install Gentoo. There are many pathways
35 to streamline the installation of Gentoo. Many, for onerous
36 reasons believe that is a bad idea.
37
38 There is plenty of existing installation code that sets up
39 MBR and ext*; so that's a no brainer on how to do that. Newer
40 technologies, like btrfs are tricky.
41
42 > In my opinion, there's really 3 parts to the install
43 process, and I think it helps to distinguish > between
44 them. I think a complete installer program has to address
45 all 3, but each task could be > modularised.
46
47 > 1. The low level decisions, like disk partitioning, raid
48 and disk mirroring, filesystem choices > like ext4, btrfs,
49 zfs, or some other. For a VM, the choices here might include
50 creating a > new LVM volume or btrfs subvolume
51
52 Gentoo is not going to formally support ZFS as has been
53 stated before. However supporting ZFS by others is well
54 documented and some maverick could easily extend the
55 gentoo-base installer for a target system (after your
56 Category-1) where ZFS is installed. Just not officially
57 gentoo.
58
59 > 2. Installing system files, which is not much more than
60 untaring the stage3, and low level > system configuration
61 of make.conf settings, choice of profile, locale & timezone
62 settings, > users & passwords, networking, choice of syslog &
63 from, etc
64
65 Category-2 This is a pretty easy part to automate. Many
66 have stated that all of this information could be gathered
67 up before the actual installation (batched) begins and
68 parsed out at the appropriate time during the actually
69 (automated) installation.
70
71 All of Category 1 as well as some parts of Category-2 are
72 what I refer to as the base-install. After that point is
73 when you make key decisions like workstation vs server vs
74 embedded vs tablet.
75
76 > 3. Higher level system configuration to get to a finalised
77 state
78 This is the part of the traditional Gentoo handbook I do
79 agree with. This is the part of the installation where noobs
80 begin to actually learn gentoo, or at least those parts
81 necessary for routine administration and usage. This is part
82 of the handbook that is trivial for experience *nix folks
83 as most are familiar with more than one package manager or
84 software installation semantic.
85
86
87 Most of these sorts of noobs (folks that struggle with
88 maintaining a *nix system) are never going to profile
89 low level kernel code or compare one file system against
90 another, so why make it mandatory to master category 1 in
91 order to install, use and enjoy gentoo? Currently, the lack
92 of a gentoo installer is exactly that:: a blocker to noobs.
93 That's not my issue:: the devs are using 'mis-direction' here
94 to prohibit the creation of a slick-smooth-unattended-useful
95 base-install semantic for those with moderate *nix skills,
96 imho. YMMV. That's my issue. Dont belive me? Just go to
97 gentoo-dev and read the flack that MaffBlaster caught
98 on the list, merely for discussion a new installation
99 semantic. Hence the focus on 'stage-4' install code.
100
101
102 > (Of course, there's quite a bit of blurring between
103 the stages.)
104
105 Yes, understood, we're talking at a high-level of abstraction
106 here.
107
108 > I'm not so interested in 1, but gentoo really shines
109 here because there are no restrictions. > But there are
110 so many options that it makes it a big task to tackle,
111 unless you pare it down > and focus on a few typical use
112 cases like a standard desktop install.
113
114 OK, well this is where we part company. If you really believe
115 this and all we end up with is a competitive distro install
116 with buntu, what's the point? Critics are going to seize
117 on this and say see, your recruiting the idiots and ricers
118 from buntu to come to gentoo for indentured support. Like
119 them, that's not really the target of my goals. I do think
120 some will learn a sufficient amount about gentoo and using
121 *nix in general that they will stick around and benefit
122 the community.
123
124 I think a solid base-install that is automated benefits
125 the folks that already know their way around *nix and
126 they can quickly benefit form the source code build out
127 tools very, very quickly. If somebody does not know the
128 difference between C and C++, I'm not for catering to their
129 needs. But the lack of a good, automated installer for the
130 basics does run off many accomplished *nix folks who could
131 then the gentoo community tremendously, imho.
132
133 If somebody wants to build a firewall, a stealth sniffer
134 (no ether traces, a bridge, an embedded system product, a
135 workstation with openrc and lxqt or wayland, a cluster (in
136 house hardware) or a cloud (outsourced hardware) then that
137 is where we need to distinguish gentoo's installation. It
138 becomes a tool for those with expertise to rapidly install
139 the base-system and then do things you cannot do with buntu.
140
141
142 Note:: I use the term 'base' to distinguish from 'default'
143 which is the terminology chosen for gentoo's profile
144 system of choices. They are similar, but I believe that
145 a base-install is less than a default system on a given
146 architecture. This is a concept that I am still trying to
147 refine for categories of systems like tablets, embedded
148 and other kinds of minimal hardware types of systems. It's
149 a different issue for a different thread.
150
151
152 > Part 2 is where it would be good to have a standardised
153 approach, along the lines of > debian's debootstrap
154 utility. Something that takes a target directory and installs
155 all the > files needed to build a bare-bones system inside
156 it. Its actually not that difficult > to write a shell script
157 to achieve this, which is probably why there are so many >
158 posted around the interests. But something standardised
159 could be the basis of a gui > installer, or the center of
160 a container installer such as the lxc-gentoo script or >
161 whatever the docker equivalent is.
162
163 What you are talking about here, to me, can be part of
164 your category-3. In fact, I modify what you have written
165 to only (2) parts Step-1(base-install) and Step-2 (Final
166 Target). Step-2 does not need to be automated by gentoo
167 officially imho. It's mostly a few config questions and
168 a collect of packages. You can actually auto this right
169 now. Just create your own 'profile' on top of a default
170 installation and use ansible.
171
172 MuffBalster said the first part is the easiest part;
173 that is to go back to supporting (stage-4) installs. Lot
174 of subtle intricacies with 'stage-4' but I'll give you my
175 take so you can see where the devs are rolling.
176
177 If you have expertise and have created one utopian install
178 for a highly targeted need, then a stage-4 installation
179 semantic will allow you to quickly clone that work and build
180 many more similar system, in a fully automated fashion. So
181 for Clustering, cloud, google, large institutions or
182 consultants with advanced skills, it become a defacto private
183 installation sematic where the inner circle and those with
184 a strong set of admin/programming skills prosper. Maybe an
185 inner circle? The noobs are still filtered out. Those in
186 the middle to do gain from the knowledge and expertise of
187 the inner core of devs. Is that the way it should be? Not
188 for me to decide. I'd just like some help with newer file
189 systems (btrfs), gpt, grub2, and such issues in Step-1.
190
191 Those that have painted my position to be recruiting the
192 masses of noobs from buntu, are using mis-direction to
193 prevent the large number of moderate-skilled gentoo community
194 from the benefits of a streamlined installation. The
195 fact is that gentoo had a very useful, mostly automated
196 installation, and it was abandoned because of the catering
197 to the noob issues. But they did not have to abandon the
198 STEP-1 needs of those with sufficient *nix skills to benefit
199 from installation automation.
200
201 > The 3rd task is more in the realm of tools such as ansible
202 or puppet.
203
204
205 Step 3 already exists in rudimentary form::
206
207 Both Sephan and Alan have posted on using ansible for
208 gentoo installs. It's just not polished to the point of
209 being an installer, imho. There are many other formulas,
210 playbooks and such for installing gentoo, but they are
211 cumbersome, imho. Folks stop short of finishing them,
212 as is their right to do so. Once you go down the path of
213 applying noob_polish to the installation semantic, it does
214 become an unsustainable nightmare, or at least that is the
215 propaganda:: which I am neutral on.
216
217
218 >> Having that said, and having done few Gentoo
219 installations: I'm merely >> wishing installing Gentoo
220 wasn't such a lengthy process. It's lengthy >> in that you
221 have to do the steps manually while browsing the excellent
222 >> handbook.
223
224 Ah, so you agree, streamline the base and most can streamline
225 the target part of their installation needs without much
226 assistance; thus avoiding noob_population_explosion.
227
228 >> If there was an installer that would guide you through
229 these steps and >> bring up the files you need to edit
230 in an editor, that could save a lot >> of time already.
231 It could reduce the possibility for error, as in >>
232 overlooking that you need to do some step.
233
234
235 > Which is what part 2 is about. I started writing
236 my own installer based on using > ebuild files for the
237 configuration. But I like your idea of an interactive mode
238 for configuration.
239
240 >> Otoh, I have to come to like how Gentoo is installed.
241 You can do >> whatever you like, and the process is pretty
242 straightforward. I don't >> see how an installer could
243 give you that, yet a perfect installer would >> need to.
244
245 > And that is the difficulty inherent in a gentoo
246 installer... If its too restrictive, > its not really gentoo
247 anymore; if its flexible to cover all the options, you may
248 as > well just stick with typing commands in a shell...
249
250 I disagree. If the base-install is automated then the
251 second half of customizations are trivial to automate or
252 following the handbook to learn gentoo, as it is very easy
253 and straight-forward. Once you do that part a few times,
254 it's a collection of configs and ebuilds that are easy to
255 master and automate.
256
257 >> How about support for booting from ZFS? I'd really like
258 to see that; >> itshould be as easy as booting from other
259 file systems. Without it, we >> have to do ugly things.
260
261
262 ZFS is a different (license?) issue. There was a ZFS livedvd
263 circa 2012..... Go look at that for ideas.
264
265
266 If I seem 'conciliatory' to those opposed to the install
267 software development, it's because I put in some time
268 looking at the issue from their prospective. Your breaking
269 apart of the installation ssemantics did help me focus on
270 the Step-1 issue; which is the point and what need to be
271 automated, imho.
272
273
274 Last Zinger:: SystemD created quit a stir on this list right
275 down to the heart of gentoo's soul as a distro. I believe
276 that was healthy. When I go round the net, that is still
277 the single biggest issue where gentoo get's respect from
278 all corners (except Lennart's hottub parties)......
279
280 I just which we, gentoo as a distro, were embracing and
281 automating the installation of BTRFS, like we pursued
282 systemd. BTRFS is a game changer and many advanced folks
283 and institutions are using it. It's a bear. So, I'd settle
284 (be very happy) if out of these installer threads we just
285 end up with an automated way to complete a BTRFS, raid-1
286 system that includes a completed fstab, gpt and grub2
287 issues in some sane and automated way. WE can still filter
288 out the noobs........OK?
289
290
291
292 PEACE?
293 James

Replies

Subject Author
Re: [gentoo-user] Re: Project:Installer Bruce Schultz <brulzki@×××××.com>