FAQ Search Today's Posts Mark Forums Read
» Video Reviews

» Linux Archive

Linux-archive is a website aiming to archive linux email lists and to make them easily accessible for linux users/developers.


» Sponsor

» Partners

» Sponsor

Go Back   Linux Archive > Gentoo > Gentoo Desktop

 
 
LinkBack Thread Tools
 
Old 09-12-2010, 01:58 AM
Lindsay Haisley
 
Default Desktop problem with /dev/hda

On Sun, 2010-09-12 at 00:52 +0000, Duncan wrote:
> Lindsay Haisley posted on Sat, 11 Sep 2010 18:36:21 -0500 as excerpted:
> > Ugh!! This sounds like the roadblock I ran into! The problem drive,
> > /dev/hda, holds the root filesystem and several others, and didn't show
> > up at all in /dev. Since the system also has several SATA drives, I'm
> > using the sata_promise kernel module for these, but the SATA system on
> > the MB is managed by a Promise chipset, which supposedly implements
> > hardware RAID but it's proprietary so I'm just using plain old discrete
> > non-RAID mode.
>
> JBOD (just a bunch of disks) mode. I use it here, too.

JBOD - That's like POTS in telco terminology :-) I was trying to think
of JBOD but couldn't remember it.

> From what you wrote earlier, I thought you had all SATA hard drives but
> were depending on udev's compatibility rules to setup hdX symlinks to the
> sdX devices, using the hdX symlinks in your fstab. If you're still using
> pata and the devices themselves were hdX but are now sdX, that's a
> different issue and there's a bit more excuse...

No, the MB has an Intel chipset with a standard PATA header/port, but it
also has an onboard Promise controller and 4 SATA connectors intended to
support RAID. I never could get the Linux hacks with the proprietary
Promise RAID controller to work properly, so I just set them up JBOD and
incorporated them into Linux RAID-1 arrays.

This isn't as bad as our firewall/file server box! I have that set up
with EVMS, and it boots from an initrd right into EVMS, so the boot
drive is a Linux RAID array. Now EVMS has been abandoned (pity, it was
a great idea!) and the box is living on borrowed time as far as upgrades
are concerned. It was a daring trick, but it'll probably bite us in the
butt eventually!

> tho as mentioned really
> only if you've been living in a cave or under a rock for some years.

I like caves, and under rocks, too ;-) I have a life outside of
geek-land and these days it's looking pretty cool. I do fall behind on
this stuff, though, as a result.

> If you configure your own kernel, you should have run into that when doing
> the make oldconfig, and could have adjusted accordingly then. But if you
> depend on genkernel... well, let's just say I like to know what changes
> are going on with my kernel, and that's one of the reasons I don't use
> genkernel. (Tho for all I know, there was a warning when genkernel did
> the change too, but I wouldn't know, as I don't use it.)

I don't use genkernel either, for the same reasons. I've been
configuring my own kernels since Linux kernel 1.something. That kinda
dates me, I guess. The only thing genkernel is good for is if you
_have_ to build an initrd, but I had to do that by hand anyway when I
set up our in-house file server because I had to build EVMS support into
it.

> > I have rather a conflict here, since I already have a /dev/sda.
>
> See vvv (below, those are arrows).
>
> > Is there a HOWTO for using libata-supported kernel components, and
> > configuring them in the kernel?
>
> Someone familiar with the specific hardware you have might know exactly
> what order (vvv) the devices would appear in, but I'm not, so it's of no
> significance to me and I snipped it. I explain the general situation vvv.

Duncan, thanks for your notes and observations. Most of this I already
know, but it's always encouraging and helpful to have someone else lay
it out. I'll probably take another run at the upgrade when my time
permits, which it doesn't right now. I just need to make sure that I
have a path to back out of any changes, as I did earlier this week, if
things get foobar. I have to be able to get to a stable, bootable
desktop at the end of the day, whether it's running the old kernel or a
new one.

> Among other things, udev should create symlinks for each device UUID/GUID
> to the associated name, and if you write down which GUID applies to which
> device on your current system, you can use that to figure out which sdX
> they end up on with the new layout.

I can also look at them with cfdisk and figure it out from the partition
layout - which is probably easier. I would assume the SATA drives will
be in the same order. The hard drives are also identified by model and
what I assume to be the serial number in /dev/disk/by-id - e.g.
"ata-WDC_WD360GD-00FNA0_WD-WMAH91500602" - which is, in a pinch, also
printed on the drive cases.

> Once you've noted the order and figured out which sdX corresponds to which
> device, make your rootfs writable as you did before, and change the
> corresponding fstab and intermediate layer (lvm/dmraid/mdraid/etc) configs
> so the mapping is to the new device names instead of the old ones.

grep -R hd /etc/lvm/* says that the only place that "hd" is mentioned
(other than in comments) is in /etc/lvm/cache/.cache, which I would
assume I could erase and the LVM system would re-create it. "sd" isn't
in there at all. I'll also have to edit /etc/mdadm.conf which makes
specific references to /dev/sd* partitions by name.

> can then reboot, and it should come up as normal. =:^) (If it doesn't,
> there's probably a mapping you forgot to change, somewhere.)

... and if it fails, I go and cry myself to sleep and work on it the
next day ;-)

--
Lindsay Haisley | "Humor will get you through times of no humor
FMP Computer Services | better than no humor will get you through
512-259-1190 | times of humor."
http://www.fmp.com | - Butch Hancock
 
Old 09-12-2010, 03:17 AM
Dale
 
Default Desktop problem with /dev/hda

Brent Busby wrote:

On Sat, 11 Sep 2010, Lindsay Haisley wrote:


Ugh!! This sounds like the roadblock I ran into! The problem
drive, /dev/hda, holds the root filesystem and several others, and
didn't show up at all in /dev. Since the system also has several SATA
drives, I'm using the sata_promise kernel module for these, but the SATA
system on the MB is managed by a Promise chipset, which supposedly
implements hardware RAID but it's proprietary so I'm just using plain
old discrete non-RAID mode.

I have rather a conflict here, since I already have a /dev/sda.

Is there a HOWTO for using libata-supported kernel components, and
configuring them in the kernel? The MB has an ICH5 controller hub,
which I assume handles the PATA IF. lspci shows:

IDE interface: Intel Corporation 82801EB/ER (ICH5/ICH5R) IDE
Controller (rev 02)


The linux kernel here already has CONFIG_ATA_PIIX, which supposedly
talks the lingo of the ICH series I/O controller hub. What else do I
need here, or where can I go to learn more?


If you go through the list of drivers in the Sata section, toward the
second half of the list, there are Pata drivers. They used to be
deprecated, but now they're the only ones that are supported. You
just need to find the Pata drivers (in the Sata section) that
correspond to what you've got (you'll probably have to look around a
bit), and then turn off the entire old Pata section. There should
probably be a Pata driver in the new Sata section though that handles
Intel PIIX chipset; it's pretty common.


The danger that will make you want to keep a rescue CD handy is that
it's hard to predict how your devices are going to get enumerated at
bootup under these new semantics, or to know whether it will match
what you've got in your fstab. If you need a good rescue CD though,
I'd recommend System Rescue CD (http://www.sysresccd.org/), which is
based on Gentoo and handles about anything.




Noticed something in another reply that I thought I would mention right
quick. My mobo has on board IDE controller. I also have a SATA PCI
card with a drive connected to it as well. This may help you to know.
When I recently converted mine over to PATA/SATA, it sees the drive on
the SATA controller first then the drives connected to the mobo. So,
you may want to anticipate that the SATA drives will be sda, sdb and so
on, then any drives connected to your mobo afterwards. This was
somewhat of a surprise when I did this. Thank goodness grub is pretty
forgiving and helps the idiot in the chair. lol I wasn't sure where
my root partition was anymore.


Hope that helps just in case you have something similar.

Dale

:-) :-)
 
Old 09-12-2010, 09:02 AM
Duncan
 
Default Desktop problem with /dev/hda

Lindsay Haisley posted on Sat, 11 Sep 2010 20:58:32 -0500 as excerpted:

>> JBOD (just a bunch of disks) mode. I use it here, too.
>
> JBOD - That's like POTS in telco terminology :-) I was trying to think
> of JBOD but couldn't remember it.

Yes. I think of the two (POTS and JBOD) together as well. Too bad not
all acronyms are that simple and sensible! (I never /did/ figure out the
various SCSI versions, and pretty much everyone agrees that USB speeds are
screwed up, due to marketing, the reason nobody actually /uses/
full/high/super/whatever-speed, in reference to USB, but instead uses
USB-1/2/3.) =:^(

--
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
 
Old 09-12-2010, 10:12 AM
Duncan
 
Default Desktop problem with /dev/hda

Lindsay Haisley posted on Sat, 11 Sep 2010 19:17:36 -0500 as excerpted:

> On Sat, 2010-09-11 at 23:13 +0000, Duncan wrote:

>> I wish there were some way to really drum this into every Gentoo user's
>> head when they started, so they never ended up having to learn it the
>> hard way, as you did. But as they say, if wishes were fishes...
>
> You know, I set up my Gentoo boxes - 2 commercial servers and a desktop
> box - over 5 years ago.

I started in 2004. What was ironic was that for some reason I never did
actually figure out, 2004.0 didn't work for me, and by the time I got
around to working on it again, 2004.1 was out (in those days Gentoo did
four releases a year, one a quarter). But in the mean time I was
following the user list, the dev list, the desktop list, and the amd64
list. I had also read the handbook over, including the working with
portage and working with Gentoo sections (there wasn't yet a network
section). Plus, I read into the list archives a bit. So I had a decent
feel for all the common places folks had problems, and was actually
answering questions about Gentoo and helping people with the common
problems anyway, from my still-Mandrake box, before I even had Gentoo up
and running!

I always thought it was a shame how many folks read only the install
section of the handbook, and that only once, when they were actually
installing. Those folks may get a Gentoo system up and running, but miss
all the good hints that make it easier to administer! =:^( I wish there
were some way to have everybody go thru the process I did, actually
reading the handbook, then helping out on the forms/lists/irc, whichever
they prefer, for at least a month, before they actually got a working
install up and running. It'd make people's experience a /lot/ smoother,
once they did get up and running.

> Gentoo was a lot simpler then. There were no
> eselect news items to read because there was no eselect news. There was
> also no decent system to read the emerge notes, so Eldad Zack and I
> wrote one.

I think comparing it to steering systems is reasonable.

I installed from stage-1, because I wanted to understand it from the
ground up. And remember the bootstrapping script? Due to hardware issues
(borderline memory, really wasn't up to the clocking it was rated at, and
my system BIOS didn't have a way to underclock it until a BIOS update some
time later, after which it was solid as a rock on the same memory... until
I upgraded memory some years later), I couldn't get that whole script to
run at once, so I opened it up in an editor and did each step of the
script manually, redoing it, sometimes crashing and rebooting, until the
step completed successfully, after which I went to the next step.

I'd compare that to "tracked vehicle steering". Like Caterpillars or
other tracked vehicles, where you steer with levers that stop the tracks
on one side or the other, while the other continues to turn, so the
vehicle turns toward the stopped side.

Conventional stage-1 or stage-2 installs, like we did back then, without
eselect news, etc, would be like conventional direct steering, while using
the installer that was available for awhile, or a stage-3 tarball, is like
power steering.

Each level is easier, but more complex and farther removed from the
details. Power steering is a lot easier than conventional steering, which
is easier than tracked vehicle steering, but you lose the feel for the
road -- the reason sports-car enthusiasts generally prefer manual steering
and gearing -- automatic/power takes all the fun out of it!

> I've had to learn a lot of stuff "the hard way" but as the
> years go by I have less and less patience with having to get under the
> hood and tinker. I'm kinda stuck with what I have. My servers are
> running mysql 5.0. If I took them off line to upgrade everything to
> mysql 5.1, every system on the boxes on which my customers depend would
> break - mail service, DNS, SpamAssassin, billing, to mention a few, and
> these would be down for who knows how long. I'm almost 70 years old.
> I'll probably sell my business and let someone else worry about this
> crap before I get everything truly up-to-date.
>
> As far as the desktop system goes, I'll probably build a 2nd box, maybe
> running another distribution, and migrate stuff to it incrementally. My
> time and my sanity have value, and doing this may be less costly, in the
> long run, than trying to hack this 5-year old box and being without a
> desktop (and my company's billing system, and my email, and my web
> development tools, etc. etc.) until I get it figured out.

Wow! I'm often one of the older guys around, both in Linux dev circles
(I'm not really a dev but I enjoy hanging out with them and speaking the
lingo), where so many are in college, and quit when they get done and take
normal employment, and in my regular non-computer-related job. But I'm
only in my lower to mid 40s (nearing 44).

I've often wondered how long I'll keep up with Linux and Gentoo... tho I
do find Gentoo a perfect match for me right now. I've been around Gentoo
for over six years now, and expect that if it's still around in updated
but reasonably similar purposed form a decade from now, I'll very likely
still be running it. Two decades... it's very tough to predict /what/
computers will be like 20 years out, and Gentoo could easily be long gone
history by then, or changed so much it wouldn't be recognizable, but
still, it's conceivable that if it's still around, I might still be
running it. Put it this way: I don't foresee a reason to change,
assuming Gentoo's still around in similar purposed form, by then.

70's a bit beyond that, for me. Hopefully I'm still in reasonable shape
by then. I've idly speculated what it might be like when my generation
gets to that age. We're really the first ones to have computers, at least
C64 level, as kids or teens. How will that affect our approach to
technology as we age? I really don't know, but I sort of have this
picture in my head of me being involved with and perhaps president of the
LUG in my retirement home! =;^)

Would I still have the patience to run Gentoo, or would I be running
something really simple and hand-holdy, like Ubuntu, by then? What
questions to be contemplating! =:^)

FWIW, I've read very good things about Arch, including from a number of
former Gentooers who got tired of the full from-source for /everything/.
Apparently, it allows a lot more control of the installation than most
binary distributions, with rather less hassle than Gentoo. Like Gentoo,
it's a rolling distribution, something I'd consider a bonus. If I were to
consider taking it down a notch, that would be the first one I'd try. So
that's what I'd suggest, if indeed you are considering taking it down a
notch.

Beyond that, I think Debian unstable would be my next choice, for the
desktop, probably testing for servers. They're big enough to have the
power of numbers behind them, both people and packages, and are a
community distribution. Here at least, I consider that a good thing.
I've tried commercial/company-backed distributions and simply don't find
them appropriate for me. As such, I doubt I'll ever run a Mandriva,
Fedora, or Ubuntu, again, unless it happens due to my switching to the
computer field for my job, and it ends up just being simpler to run the
same thing on my own computers as well. But I don't see myself as happy
enough with such distributions to ever run them on my own. Thus, if I
/were/ to go mainstream binary distribution, Debian is almost certainly
what I'd choose, over the Redhats/Fedoras, NLSs/NLDs/SuSEs/OpenSuSEs,
Mandrivas, Ubuntus, etc.

> Enough of this geezer-rant. Peace and love to everyone in these crazy
> times. I mean it!

--
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
 

Thread Tools




All times are GMT. The time now is 07:31 AM.

VBulletin, Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Content Relevant URLs by vBSEO ©2007, Crawlability, Inc.
Copyright 2007 - 2008, www.linux-archive.org