Hello,

I have recently installed Slackware 10.0. I am using a Promise's
Fasttrak66 RAID controller on the computer, (hopefully) making two
identical disks to appear for Linux as a single drive.

The installation from the CDs went quite well, there wasn't any
errors. At first there was some make errors and learn, but after I
figured out I have to use the ataraid.i -kernel instead of bare.i.

The mirrored drive was also now found as /dev/ataraid/d0, instead of
/dev/hdxx.

Installation wasn't hard, but then I tried to start the freshly
installed system. It hangs up after lilo has loaded (notice, loading
lilo from mirrored disks appear to work....) and I have selected the
Linux to boot. It starts booting into OS, but hangs up almost
immediately, where it reads on screen "Booting Linux.........".

The dots after text come dot by dot on screen, and hangs up always the
same position, about 10-15 dots. I can check the exact number of dots
it it has any meaning.

Does anyone know why it hangs, and how to get it working? If not, is
there any way to find out what makes the starting OS to hang?

JD

> Linux to boot. It starts booting into OS, but hangs up almost
> immediately, where it reads on screen "Booting Linux.........".
>
> The dots after text come dot by dot on screen, and hangs up always the
> same position, about 10-15 dots. I can check the exact number of dots
> it it has any meaning.

Don't the dots represent the kernel image being uncompressed in memory? I
might be wrong here, but that's what I thought. The bzip compressed kernel
can uncompress to become quite large. If the compressed image is damaged or
corrupted (boot sector issue), decompression would fail.

--
Jem Berkes
http://www.sysdesign.ca/

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

In alt.os.linux.slackware, JD dared to utter,
> I have recently installed Slackware 10.0. I am using a Promise's
> Fasttrak66 RAID controller on the computer, (hopefully) making two
> identical disks to appear for Linux as a single drive.

Ick. I've got one of those two and while I am using it, I'm not at all
fond of it. Performance is very low for one, and support for that
chipset is flakey. Basically it's software RAID, not hardware RAID;
even though there's a chip there, the CPU does pretty much all the
work. FWIW, I believe other people have had much better experiences
using LVM and Linux's native software RAID. I would be using that, but
I managed to get this machine working and don't want to fool with it
anymore. :^)

> Installation wasn't hard, but then I tried to start the freshly
> installed system. It hangs up after lilo has loaded

To trouble shoot this, why don't you see if you can boot into your
install using the ataraid.i kernel on the installation cd or live cd?
Instructions on how to do that are provided as soon as the disk boots.

- --
It is better to hear the rebuke of the wise,
Than for a man to hear the song of fools.
Ecclesiastes 7:5
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)

iD8DBQFBK/WxlKR45I6cfKARAvp1AKC0SJYuPBvC51dfCVwEXbIw9zZXdgCg lVaW
yKp5ndydF4MQP+u21Fqzmbk=
=sGl7
-----END PGP SIGNATURE-----

NOTE: This message was sent thru a mail2news gateway.
No effort was made to verify the identity of the sender.
--------------------------------------------------------

pgp trash troll delete

Hicks, Alan
188 Shady Dale Dr
Lizella, GA 31052
478-935-8132

+Alan Hicks+ <alan@lizella.network> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1

> In alt.os.linux.slackware, JD dared to utter,
> > I have recently installed Slackware 10.0. I am using a Promise's
> > Fasttrak66 RAID controller on the computer, (hopefully) making two
> > identical disks to appear for Linux as a single drive.

> Ick. I've got one of those two and while I am using it, I'm not at all
> fond of it. Performance is very low for one, and support for that
> chipset is flakey. Basically it's software RAID, not hardware RAID;
> even though there's a chip there, the CPU does pretty much all the
> work. FWIW, I believe other people have had much better experiences
> using LVM and Linux's native software RAID. I would be using that, but
> I managed to get this machine working and don't want to fool with it
> anymore. :^)

> > Installation wasn't hard, but then I tried to start the freshly
> > installed system. It hangs up after lilo has loaded

> To trouble shoot this, why don't you see if you can boot into your
> install using the ataraid.i kernel on the installation cd or live cd?
> Instructions on how to do that are provided as soon as the disk boots.

> - --
> It is better to hear the rebuke of the wise,
> Than for a man to hear the song of fools.
> Ecclesiastes 7:5
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.2.1 (GNU/Linux)

> iD8DBQFBK/WxlKR45I6cfKARAvp1AKC0SJYuPBvC51dfCVwEXbIw9zZXdgCg lVaW
> yKp5ndydF4MQP+u21Fqzmbk=
> =sGl7
> -----END PGP SIGNATURE-----

Jem Berkes wrote:

>> Linux to boot. It starts booting into OS, but hangs up almost
>> immediately, where it reads on screen "Booting Linux.........".
>>
>> The dots after text come dot by dot on screen, and hangs up always the
>> same position, about 10-15 dots. I can check the exact number of dots
>> it it has any meaning.
>
> Don't the dots represent the kernel image being uncompressed in memory? I
> might be wrong here, but that's what I thought. The bzip compressed kernel
> can uncompress to become quite large. If the compressed image is damaged
> or corrupted (boot sector issue), decompression would fail.
>

This happens also if it can't find a bootable partition... ie if Lilo isn't
installed on the boot partition (so you need another bootloader to point to
lilo). Check lilo is installed on *boot.

On 24 Aug 2004 21:14:13 -0500, +Alan Hicks+ <alan@lizella.netWORK>
wrote:

>-----BEGIN PGP SIGNED MESSAGE-----
>Hash: SHA1
>
>In alt.os.linux.slackware, JD dared to utter,
>> I have recently installed Slackware 10.0. I am using a Promise's
>> Fasttrak66 RAID controller on the computer, (hopefully) making two
>> identical disks to appear for Linux as a single drive.
>
>Ick. I've got one of those two and while I am using it, I'm not at all
>fond of it. Performance is very low for one, and support for that
>chipset is flakey. Basically it's software RAID, not hardware RAID;
>even though there's a chip there, the CPU does pretty much all the
>work. FWIW, I believe other people have had much better experiences
>using LVM and Linux's native software RAID. I would be using that, but
>I managed to get this machine working and don't want to fool with it
>anymore. :^)
>

Thanks for information about the card. It indeed looks like this card
doesn't have very good support under Linux, I am seriously thinking
about discarding the card (it is a shame, I bought it used just for
this project..) and just do the RAID by Linux's own tools.

It also forces me to rearrange the disks. While there is now 7 disk
devices in computer due to cards extra interfaces, I have to get it
working with 4 (or buy yet another IDE-controller...)

Do you know, is there any option to install the Slackware 10 so the
RAID (I prefer mirroring in this case) would be in use right from the
start?

Is it possible to have the installation done so the mirrored disks are
in the end truly identical - it really doesnt sound a good idea if I
have to modify Lilo menu so there is 2 different startup choices, one
for /dev/hda and other for /dev/hdc for example...



>> Installation wasn't hard, but then I tried to start the freshly
>> installed system. It hangs up after lilo has loaded
>
>To trouble shoot this, why don't you see if you can boot into your
>install using the ataraid.i kernel on the installation cd or live cd?
>Instructions on how to do that are provided as soon as the disk boots.

Tried this, and it helped a bit. The boot still hangs, but maybe
helpfully I got the last screenful of text to see some error messages:


hdg: hdg1 hdg2
ataraid/d0:<1>Unable to handle kernel NULL pointer dereference at
virtual address 00000000
printing eip:
00000000
*pde = 00000000
Oops: 0000
CPU: 0
EIP: 0010:[<00000000>] Not tainted
EFLAGS: 00010286
<some 8 digit hex values and their keys>
Prosess swapper (pid: 1, stackpage=d3ff5000)
Stack: <some 8 digit hex values>
Call Trace: <some 8 hex values>

Code: Bad EIP value.
<0>Kernel panic: Attempted to kill init!


Doesn't really say much to me, but as I see this, it tries to say
there is some problem with that ataraid/d0 disk (or what I believe,
the controller..).

JD

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

In alt.os.linux.slackware, JD dared to utter,
> Thanks for information about the card. It indeed looks like this card
> doesn't have very good support under Linux, I am seriously thinking
> about discarding the card (it is a shame, I bought it used just for
> this project..) and just do the RAID by Linux's own tools.

That's up to you. You should have got a 3ware controller. Those are
well supported by pretty much all OS's and perform much better than
the broken promise[0] controller.

> It also forces me to rearrange the disks. While there is now 7 disk
> devices in computer due to cards extra interfaces, I have to get it
> working with 4 (or buy yet another IDE-controller...)

Whoa.... that's a ton of drives. Sounds like you've got JBOD[1] and
want to turn that into a RAID of some kind, right?

> Do you know, is there any option to install the Slackware 10 so the
> RAID (I prefer mirroring in this case) would be in use right from the
> start?

Yeah. That's the way I did it. How did you goof that up? Basically you
boot the ataraid.i kernel and go from there. You partition
/dev/ataraid/d0 and run with it from there. You may have to tell the
installation that your linux partitions ain't /dev/hdg2[2] but rather
/dev/ataraid/d0p2.

> Is it possible to have the installation done so the mirrored disks are
> in the end truly identical - it really doesnt sound a good idea if I
> have to modify Lilo menu so there is 2 different startup choices, one
> for /dev/hda and other for /dev/hdc for example...

Yes of course, provided the drives are identical in size and crap like
that that matters for RAID. Just do as I said above. You'll have to set
the RAID array up first in the broken promise controller's BIOS.

> Tried this, and it helped a bit. The boot still hangs, but maybe
> helpfully I got the last screenful of text to see some error messages:
>
>
> hdg: hdg1 hdg2
> ataraid/d0:<1>Unable to handle kernel NULL pointer dereference at
> virtual address 00000000
> printing eip:
> 00000000
> *pde = 00000000
> Oops: 0000
> CPU: 0
> EIP: 0010:[<00000000>] Not tainted
> EFLAGS: 00010286
><some 8 digit hex values and their keys>
> Prosess swapper (pid: 1, stackpage=d3ff5000)
> Stack: <some 8 digit hex values>
> Call Trace: <some 8 hex values>
>
> Code: Bad EIP value.
> <0>Kernel panic: Attempted to kill init!

Maybe something here will help?

http://www.google.com/search?q=slack...UTF-8&oe=UTF-8

> Doesn't really say much to me, but as I see this, it tries to say
> there is some problem with that ataraid/d0 disk (or what I believe,
> the controller..).

Could it be that you SNAFU'd the installation and the /etc/fstab file
has calls to /dev/hdXX instead of /dev/ataraid/d0XX ?

[0] Littler play on words there. :-)
[1] Just a Bunch Of Disks. Yes, that really is a computer industry
acronym.
[2] Entirely contrived example of course.

- --
It is better to hear the rebuke of the wise,
Than for a man to hear the song of fools.
Ecclesiastes 7:5
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)

iD8DBQFBLkL2lKR45I6cfKARAjPxAJ0YhdHjgm7MZoiranjPjx h0zb4kVQCfY7GK
m9aqj87bldGvYpHtLhLg99I=
=79Op
-----END PGP SIGNATURE-----

NOTE: This message was sent thru a mail2news gateway.
No effort was made to verify the identity of the sender.
--------------------------------------------------------

pgp trash troll delete

Hicks, Alan
188 Shady Dale Dr
Lizella, GA 31052
478-935-8132

+Alan Hicks+ <alan@lizella.network> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1

> In alt.os.linux.slackware, JD dared to utter,
> > Thanks for information about the card. It indeed looks like this card
> > doesn't have very good support under Linux, I am seriously thinking
> > about discarding the card (it is a shame, I bought it used just for
> > this project..) and just do the RAID by Linux's own tools.

> That's up to you. You should have got a 3ware controller. Those are
> well supported by pretty much all OS's and perform much better than
> the broken promise[0] controller.

> > It also forces me to rearrange the disks. While there is now 7 disk
> > devices in computer due to cards extra interfaces, I have to get it
> > working with 4 (or buy yet another IDE-controller...)

> Whoa.... that's a ton of drives. Sounds like you've got JBOD[1] and
> want to turn that into a RAID of some kind, right?

> > Do you know, is there any option to install the Slackware 10 so the
> > RAID (I prefer mirroring in this case) would be in use right from the
> > start?

> Yeah. That's the way I did it. How did you goof that up? Basically you
> boot the ataraid.i kernel and go from there. You partition
> /dev/ataraid/d0 and run with it from there. You may have to tell the
> installation that your linux partitions ain't /dev/hdg2[2] but rather
> /dev/ataraid/d0p2.

> > Is it possible to have the installation done so the mirrored disks are
> > in the end truly identical - it really doesnt sound a good idea if I
> > have to modify Lilo menu so there is 2 different startup choices, one
> > for /dev/hda and other for /dev/hdc for example...

> Yes of course, provided the drives are identical in size and crap like
> that that matters for RAID. Just do as I said above. You'll have to set
> the RAID array up first in the broken promise controller's BIOS.

> > Tried this, and it helped a bit. The boot still hangs, but maybe
> > helpfully I got the last screenful of text to see some error messages:
> >
> >
> > hdg: hdg1 hdg2
> > ataraid/d0:<1>Unable to handle kernel NULL pointer dereference at
> > virtual address 00000000
> > printing eip:
> > 00000000
> > *pde = 00000000
> > Oops: 0000
> > CPU: 0
> > EIP: 0010:[<00000000>] Not tainted
> > EFLAGS: 00010286
> ><some 8 digit hex values and their keys>
> > Prosess swapper (pid: 1, stackpage=d3ff5000)
> > Stack: <some 8 digit hex values>
> > Call Trace: <some 8 hex values>
> >
> > Code: Bad EIP value.
> > <0>Kernel panic: Attempted to kill init!

> Maybe something here will help?

> http://www.google.com/search?q=slack...UTF-8&oe=UTF-8

> > Doesn't really say much to me, but as I see this, it tries to say
> > there is some problem with that ataraid/d0 disk (or what I believe,
> > the controller..).

> Could it be that you SNAFU'd the installation and the /etc/fstab file
> has calls to /dev/hdXX instead of /dev/ataraid/d0XX ?

> [0] Littler play on words there. :-)
> [1] Just a Bunch Of Disks. Yes, that really is a computer industry
> acronym.
> [2] Entirely contrived example of course.

> - --
> It is better to hear the rebuke of the wise,
> Than for a man to hear the song of fools.
> Ecclesiastes 7:5
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.2.1 (GNU/Linux)

> iD8DBQFBLkL2lKR45I6cfKARAjPxAJ0YhdHjgm7MZoiranjPjx h0zb4kVQCfY7GK
> m9aqj87bldGvYpHtLhLg99I=
> =79Op
> -----END PGP SIGNATURE-----

On 26 Aug 2004 15:08:25 -0500, +Alan Hicks+ <alan@lizella.netWORK>
wrote:

>-----BEGIN PGP SIGNED MESSAGE-----
>Hash: SHA1
>
>In alt.os.linux.slackware, JD dared to utter,
>> Thanks for information about the card. It indeed looks like this card
>> doesn't have very good support under Linux, I am seriously thinking
>> about discarding the card (it is a shame, I bought it used just for
>> this project..) and just do the RAID by Linux's own tools.
>
>That's up to you. You should have got a 3ware controller. Those are
>well supported by pretty much all OS's and perform much better than
>the broken promise[0] controller.
>

I agree on this, read on...

>> It also forces me to rearrange the disks. While there is now 7 disk
>> devices in computer due to cards extra interfaces, I have to get it
>> working with 4 (or buy yet another IDE-controller...)
>
>Whoa.... that's a ton of drives. Sounds like you've got JBOD[1] and
>want to turn that into a RAID of some kind, right?
>

Yes, something like that :-) The idea is just to have some portion of
disks to be mirrored and failsafe area, and then just utilize the all
kinds of IDE disks left useless in shelf as a bit bigger storage.
Quite trivial isn't it. It really gives some possibilities when there
is more drives than just the system disks.

>Maybe something here will help?
>
>http://www.google.com/search?q=slack...UTF-8&oe=UTF-8
>

Yes. Very much help from this link - made my decision to just forget
this controller, there isn't really any advantage of using it with
Linux. Moreover, looks like it brings just troubles instead of just
getting that RAID working....

It doesn't really sound very robust solution if it is Windows-only.
Better find some other card, having supported for other OS's too.

Many thanks for helping.

JD