Gentoo Archives: gentoo-server

From: "Benjamen R. Meyer" <bm_witness@×××××.com>
To: gentoo-server@l.g.o
Subject: Re: [gentoo-server] Challenging Update Question
Date: Tue, 12 Feb 2008 13:16:23
Message-Id: 47B19B91.2050101@yahoo.com
In Reply to: [gentoo-server] Challenging Update Question by Collin Starkweather
1 How about option #3: Build a hard drive (or hard drive set) locally for
2 the system and ship them to be installed?
3
4 I think that would give you the best of both of your previous options.
5 Granted, you'd have to go through export compliance stuff and all that,
6 but it might be worth the extra hassle.
7
8 Just a thought.
9
10 Ben
11
12 Collin Starkweather wrote:
13 > I have a server that is in need of a significant update, but it's
14 > proving a challenge. I have a big picture question, then provide some
15 > details below.
16 >
17 > (I originally thought the gentoo-admin list would be the best place to
18 > ask this, but based on the stats, it seems to be relatively inactive.
19 > Let me know if there is a better list to ask the question on.)
20 >
21 > Please excuse the length of the question, but as you can see, there are
22 > a variety of variables in play.
23 >
24 > The Big Picture
25 > ---------------
26 >
27 > The server has not been updated since late 2005 or so. It just runs
28 > Apache, mod_perl, and an application server. So far, it has just hummed
29 > along doing its work without complaint, solid as a rock, which is why no
30 > one has bothered with it.
31 >
32 > As you doubtless know, if you miss a couple of upgrade cycles with
33 > Gentoo, there can be (and has been) breakage when trying to emerge -u
34 > world.
35 >
36 > There are two identical drives, and I've mirrored (manually, not RAID)
37 > one onto the other.
38 >
39 > The challenge is that the box is in the U.S. and I live in China. There
40 > is no one there who can administer it; however, if something goes really
41 > wrong, someone can just swap the drives and reboot. The key factor is
42 > that I want this to be as low-risk as possible since swapping the drives
43 > is about the extent of on-site support available.
44 >
45 > The big picture question: Would it (1) be simpler and easier to rebuild
46 > from scratch on the redundant drive, or (2) is it simpler and easier to
47 > deal with the current issues updating from 2005.0 and a 2.4.x kernel?
48 >
49 > The Details
50 > -----------
51 >
52 > Option (1): Rebuilding on the Redundant Drive
53 >
54 > Pros -- It seems this would be the easiest way to do things, and I get a
55 > fresh kernel and build.
56 >
57 > Cons -- If I rebuild on the redundant drive, I lose the ability to swap
58 > drives if there is breakage. Also, the application server (Apache
59 > Pagekit) is solid as a rock, but a real bitch to configure. Last time I
60 > tried to upgrade Pagekit, due to Apache versioning issues, configuration
61 > changes, etc., it took me a full weekend. Not fun.
62 >
63 > Option (2): Upgrading from 2005.0
64 >
65 > Pros -- Perhaps less risky (advice on this would be appreciated!) and I
66 > maintain another drive I can use to compare configurations, selectively
67 > roll things back, etc.
68 >
69 > Cons -- The gory details. When I did an emerge --sync, the
70 > /etc/make.profile symlink broke. It used to point at
71 >
72 > /usr/portage/profiles/default-linux/x86/no-nptl/2.4
73 >
74 > which no longer exists, I suppose since the 2.4.x kernels seem to no
75 > longer be supported.
76 >
77 > So the questions that arise are:
78 >
79 > 1) Would it be less risky to upgrade from 2005.0 to 2007.0 than rebuild
80 > from scratch on the redundant drive?
81 > 2) What is no-nptl? I don't know why the old portage profile was
82 > /usr/portage/.../no-nptl. It seems to have something to do with
83 > glibc-2.4. Do I still need it? Or can I just point /etc/make.profile
84 > at /usr/portage/.../2007.0?
85 > 3) I pointed /etc/make.profile at /usr/portage/.../2007.0, and tried
86 > emerge -pu world. I was told that mail-mta/qmail no longer existed and
87 > is required by sudo (?!?) which is required by libapreq2. Perhaps
88 > mail-mta/netqmail is the new qmail? Anyway, this gives me the
89 > impression that there are deep dependencies that may have changed
90 > significantly. Is this what you would (subjectively) characterize as a
91 > bad sign?
92 > 4) What would be the best order of operations? emerge -u world, then
93 > update the kernel, or update the kernel, then emerge -u world?
94 > 5) 2008.0 is due on March 17. Is it worthwhile putting off the upgrade
95 > for 2008.0? I wouldn't want to deal with two difficult upgrades if
96 > there is breakage between 2007.0 and 2008.0.
97 >
98 > Thanks in advance,
99 >
100 > -Collin
101 >
102
103
104 --
105 gentoo-server@l.g.o mailing list