ATAPI CD-ROM not detected in system with SATA

Is this a default installation (booting from the CD and pressing enter at the
prompt), or did you pass some options at the isolinux prompt?

Those modules are in fact not present, but they are also not needed, and should
not cause the installation to fail.

tried with default settings, tried with pci=noacpi

Created an attachment (id=139)
Shows search for files when it doesn't work

Created an attachment (id=140)
I'm seeing this behaviour too.

Shows "search for files" when it is working.

Ok, those two attatchments belong to bug 8195. I have no idea how they ended up
in this one. Sorry.

still run into this issue after disabling pretty much everything I could at
install. Says it can't find common cd drive to install from.

10:35 < joem> the only thing in messages really is PCI: Ignoring BAR0-3 of IDE
controller 0000:00:1f.2
10:35 < joem> and the standard E-IDE driver loading and setting the bus speed
10:36 < joem> oh there is this in syslog:
10:36 < joem> de0: I/O resource 0x1F0-0x1F7 not free.
10:36 < joem> Sep 23 07:29:30 localhost kernel: ide0: ports already in use,
skipping probe
10:36 < joem> then the same for ide1

Created an attachment (id=221)
full dmesg output

here is a full dmesg output. At the very end is me trying out various ide
modules/

Is the ide-cd module loaded? If not, try loading it and see if that helps.

doesn't help, still don't get a device in /sys, loading all of the other ide*
modules and the sr-mod didn't do it either.

It looks like ata_piix is claiming the IDE buses, but not detecting the devices
attached to them. Perhaps a kernel issue.

(In reply to comment #11)
> It looks like ata_piix is claiming the IDE buses, but not detecting the devices
> attached to them. Perhaps a kernel issue.

IDE devices are not initialised until ide-generic is loaded or has been loaded
before.

when the ide-generic driver is loaded, that is when I get errors like ide0: I/O
resource 0x1F0-0x1F7 not free showing up in dmesg.

Created an attachment (id=245)
dmesg output from livecd boot

If it helps, here is the dmesg output from the ubuntu live cd, where ide
devices are detected and function fine.

(In reply to comment #13)
> when the ide-generic driver is loaded, that is when I get errors like ide0: I/O
> resource 0x1F0-0x1F7 not free showing up in dmesg.

Please post the exact messages. The important bit is ide1.

ide1: I/O resource 0x170-0x177 not free.
ide1: ports already in use, skipping probe

I have managed to get around this problem by turning of ide mapping for the sata
drive in the bios.

Can you access the CDROM as a SCSI device now?

ah, sorry. Yes I can access them as scsi devices once I load the ide-generic
module.

Can you please post an updated dmesg?

Created an attachment (id=415)
Updated dmesg output

here is updated dmesg output

Thanks. Do you have any remaining issues regarding this bug report?

still not quite what was going on, since it seems to be an ubuntu specific
issue(for me). As long as things work now, and other people don't seem to be
having these problems, I am good.

(In reply to comment #23)
> still not quite what was going on, since it seems to be an ubuntu specific
> issue(for me).

Do you have another Linux system where it works with the original BIOS setting
using the SATA driver?

yea, it works fine on my gentoo install and with various live cds (ubuntu
included).

Did they use the SATA driver? If they did please post a dmesg. The ubuntu live
CD certainly doesn't use SATA.

(In reply to comment #25)
> yea, it works fine on my gentoo install and with various live cds (ubuntu
> included).

Can you provide the requested dmesg output?

sorry, lost track of this bug. I am currently away for an extended period of
time, and don't have access to that box. Feel free to close since it doesn't
seem to be an issue with anybody else and I can't provide the requested info at
this point.

I have the same / similiar problem - my 2 ide cd-rom/dvd drives are not detected
in my sata-hdd powered system and therefore I cannot install warty. :\

Created an attachment (id=651)
Dmesg from Warty-Install CD

Created an attachment (id=652)
lsmod output from warty install cd

Created an attachment (id=654)
Detailed System Summary about Motherboard, Bios, PCI, etc etc

(In reply to comment #29)
> I have the same / similiar problem - my 2 ide cd-rom/dvd drives are not detected
> in my sata-hdd powered system and therefore I cannot install warty. :\

Additionally, Fedora & Gentoo both worked fine on the same system / bios setting :\

Created an attachment (id=655)
Dmesg output from SuSE 9.2 LiveCD (where the cd-drives work)

Created an attachment (id=656)
Lsmod output from SuSE 9.2 LiveCD (where the cd-drives work)

Created an attachment (id=657)
Lspci output from SuSE 9.2 LiveCD (where the cd-drives work)

Created an attachment (id=658)
uname -a output from SuSE 9.2 LiveCD (where the cd-drives work)

Created an attachment (id=659)
Dmesg output from Warty LiveCD (where the cd-drives work)

Created an attachment (id=660)
Lsmod output from Warty LiveCD (where the cd-drives work)

Created an attachment (id=661)
Lspci output from Warty LiveCD (where the cd-drives work)

It's all classic IDE. Maybe the SATA parts are discovered correctly (I don't
have any SATA device to test).
The Device Manager says it's a Intel 82801EB/ER (ICH5/ICH5R) Ultra ATA 100
Controller (Intel Mainboard).
Devices are invisible until I start ide-generic, neither ide-disk nor ata_piix
work without ide-generic.

(In reply to comment #71)
> It's all classic IDE. Maybe the SATA parts are discovered correctly (I don't
> have any SATA device to test).
> The Device Manager says it's a Intel 82801EB/ER (ICH5/ICH5R) Ultra ATA 100
> Controller (Intel Mainboard).
> Devices are invisible until I start ide-generic, neither ide-disk nor ata_piix
> work without ide-generic.

Can you try kernel 2.6.10-25.2 and/or the latest set of daily CD builds?

(In reply to comment #72)
> Can you try kernel 2.6.10-25.2 and/or the latest set of daily CD builds?

Same result. No IDE devices, neither as SCSI (through ata_piix) nor by loading
"ide-cd", "ide-disk". ide-generic makes them visible to ide-cd, etc.

While ata_piix does detect my two SATA ports, it does not detect the other two
PATA ports.

On IRC, I asked Nico to try Chuck's ata_piix.ko which should have been
equivalent to what was in -25.1 (i.e., #define ATA_ENABLE_ATAPI #undef
ATA_ENABLE_PATA), and he confirmed that it fixed his problem, while he still has
problems with -25.2. I have no idea why, since he's dealing with disks, and not
CD-ROMs, but since we're currently planning to roll that fix in post-preview, it
should at least fix that problem.

(In reply to comment #74)
> On IRC, I asked Nico to try Chuck's ata_piix.ko which should have been
> equivalent to what was in -25.1 (i.e., #define ATA_ENABLE_ATAPI #undef
> ATA_ENABLE_PATA), and he confirmed that it fixed his problem, while he still has
> problems with -25.2. I have no idea why, since he's dealing with disks, and not
> CD-ROMs, but since we're currently planning to roll that fix in post-preview, it
> should at least fix that problem.

I had done the kernel updates to -25.1 and .25.2 without even thinking... I was
so happy when the #define ATA_ENABLE_ATAPI was put in, since before that I was
having to compile a custom kernel after installing so that my SATA DVD-BURNER
would wrok. But now with those last two kernel upgrades, my drive has stopped
working as well. Didn't notice until just now when I was about to burn a DVD.
There is quite a big patch from the libata maintaner that I've found for 2.6.11.
 Is there any chance we could get that put together, so this problem won't
continue? Though I'm thinking we'll be waiting until after Hoary for that,
since Hoary is in feature freeze, right?

I'd also like to say I'm seeing this exact problem on a Dell Inspiron 6000
laptop, and the workaround posted does not work. The installer cannot detect
the DVD drive and it can't install.

Folks, we know this is a serious kernel issue and will not release Hoary with
this bug. We had to revert to this KNOWN bug just before preview, because the
fix for this bug that we had, introduced silent data corruption on some CDs. The
kernel team assure me we can nail this one. Please test ongoing kernel updates
and report success / failure.

Shouldn't this be fixed by 2.6.10-26?

linux-source-2.6.10 (2.6.10-26) hoary; urgency=low
[...]
  * Renable enable_atapi_ata-2.dpatch.

Do the daily snapshots update the installer boot kernel as well as the kernel it
eventually installs? I am trying to help test but I'm unsure if the kernel in
the installer for daily snapshot 3-12-05 is the -26 kernel as is listed in the
list, or if the -26 kernel is installed after the installer runs and the
installer still has -4.

I went back to the daily CD for 3-8-05 that had the original fix and booted,
then used the daily CD from 3-12-05 for the install. So after I get Ubuntu
installed, -26 has the CD and HDD detected fine, and when I upgraded to -27 and
rebooted it still worked fine. So it seems for me -26 and -27 fix the problem.

Great. I've seen a few success reports and confirmed it myself. Closing this bug.

Tks,
Jeff Bailey

Should this bug be fixed with the latest daily iso-images? Because apparently it isn't... I downloaded two images, one
from March 12th and another one from yesterday (14th), and both do NOT work! Which means, the setup doesn't detect my cd-
rom's and therefore a clean ubuntu installation cannot be performed and finish (as it did with array 6)... :\

jb: Do you have a working linux system on that box right now? If yes, can you
give me the output from lspci and lspci -n?

Created an attachment (id=1670)
dmesg with cd-drives working fine (using the array-6 installation-cd plus
updates)

Created an attachment (id=1671)
dmesg with cd-drives not working (booting the latest nightly cd-images and
using terminal 2 to create output)

Created an attachment (id=1672)
lscpi output for my working system

Created an attachment (id=1673)
lscpi-n output for my working system

Yuck. Looks like this bug is showing up on a slight variation of ata_piix.
Reopening bug.

I got this kind of problem after upgrading from version -25 to -27 of the
2.6.10-4 kernel - on my old laptop, a Sony Vaio Z600 (the 2.6.10-5 version has
the same problem).

As of now, the kind of kernel messages with the new kernel are:
SCSI subsystem initialized
libata version 1.10 loaded.
Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
usbcore: registered new driver usbfs
usbcore: registered new driver hub
Initializing USB Mass Storage driver...
usbcore: registered new driver usb-storage
USB Mass Storage support registered.
pivot_root: No such file or directory
/sbin/init: 428: cannot open dev/console: No such file
Kernel panic - not syncing: Attempted to kill init!

The kind of kernel messages with the old kernel were:
SCSI subsystem initialized
libata version 1.10 loaded.
ata_piix version 1.03
ata1: PATA max UDMA/33 cmd 0x1F0 ctl 0x3F6 bmdma 0xFCB0 irq 14
ata1: dev 0 cfg 49:0f00 82:7c6b 83:4108 84:4000 85:7c69 86:0008 87:4000 88:041f
ata1: dev 0 ATA, max UDMA/66, 39070080 sectors:
ata1: dev 0 configured for UDMA/33
scsi0 : ata_piix
elevator: using anticipatory as default io scheduler
  Vendor: ATA Model: TOSHIBA MK2016GA Rev: U0.3
  Type: Direct-Access ANSI SCSI revision: 05
ata2: PATA max UDMA/33 cmd 0x170 ctl 0x376 bmdma 0xFCB8 irq 15
ata2: disabling port
scsi1 : ata_piix
Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
usbcore: registered new driver usbfs
usbcore: registered new driver hub
Initializing USB Mass Storage driver...
usbcore: registered new driver usb-storage
USB Mass Storage support registered.
SCSI device sda: 39070080 512-byte hdwr sectors (20004 MB)
[etc.]

Created an attachment (id=1731)
lspci -vv for Sony Vaio Z600 where IDE drives are not detected

(In reply to comment #89)
> I got this kind of problem after upgrading from version -25 to -27 of the
> 2.6.10-4 kernel - on my old laptop, a Sony Vaio Z600 (the 2.6.10-5 version has
> the same problem).

Oy. The log snippet that you're showing is a side effect of the brief period
where the sata driver was capable of dealing with the older piix chipsets. It's
not related to the rest of this bug, though.

If you haven't done much on the box, installing with a new snapshot, or the
array 7 CD will be the simplest fix.

Otherwise, you'll need to edit /etc/mkinitrd/modules, and add three lines:

piix
ide-disk
ide-generic

Then reinstall the newer kernel. This should allow your system to boot with the
newer kernel.

Once you have the system up, you can safely remove those lines again, and future
upgrades will detect the root driver correctly.

If you have further issues related to this, please open a new bug.

Tks,
Jeff Bailey

Quick status update: Had scheduled to work on this today, but the cdimage
download claims it needs 20 hours. Will try again tomorrow.

I talked with Jeff Bailey about this yesterday and he suggested to give the
daily hoary cd of march 20th or older a try, and after downloading, burning and
booting off the cd.. it worked (again)! :)

it's still showing up these message in dmesg
ide0: I/O resource 0x1F0-0x1F7 not free.
ide0: ports already in use, skipping probe
ide1: I/O resource 0x170-0x177 not free.
ide1: ports already in use, skipping probe

.. but they do not affect the detection and file-copying proccess. I'll do a
full install using that cd later today to check whether the 2 cd-/dvd-drives
also work under a freshly installed system and post the result here too :)

Thanks to Jeff / Ubuntu team! :)

Cheers and happy easter,
-Joerg

Ok.. i did a clean install using the march 20th image, and everything worked:
installation, post-installation, working with cds with the final system etc... :)

Fabulous, thanks.

Resquiat in pacem, 1440.

Closing this bug.

*** Bug 14720 has been marked as a duplicate of this bug. ***

*** Bug 14740 has been marked as a duplicate of this bug. ***

A fitting tribute to a long-standing bug, may it stay squashed! Thanks jbailey
et al.

Sorry, folks, but I have similar behavior with the install RC dated 2005-03-30
02:49. My config:
AMD64, 1 HDD as PATA secondary master, USB CD (Sony DRX-500ULX), USB keyboard
and mouse, no ATAPI CD, no floppy, SATA onboard.

It fails to load ide-{mod,probe-mod,detect,floppy}. It does load usb-storage.
It goes through language and keyboard fine, but can't find/mount the CD it just
booted from. There is a device file that gets created and destroyed when I
plug/unplug the USB cable to the CD.

Same behavior whether or not SATA is disabled in BIOS.

I just tried it with the daily release CD for 2005-04-04, same problem
still...maybe no surprises there since the MD5SUM for that file matches that for
the RC. You guys should make sure that the dates on the dailies as listed on
the download web pages match the dates on the files when they're downloaded. I
just wasted a lot of time downloading and testing that daily, which just
duplicates the RC from last week.

--jh--

(In reply to comment #99)
> It fails to load ide-{mod,probe-mod,detect,floppy}.

Those messages are confusing but normal, and the installer only displays them to
the user in expert mode.

> I just tried it with the daily release CD for 2005-04-04, same problem
> still...maybe no surprises there since the MD5SUM for that file matches that for
> the RC. You guys should make sure that the dates on the dailies as listed on
> the download web pages match the dates on the files when they're downloaded. I
> just wasted a lot of time downloading and testing that daily, which just
> duplicates the RC from last week.

The md5sums for the release candidate are:

c8faea8b9905a07bec5f08d62f80ea37 hoary-rc-install-amd64.iso
c868ccfe958f85f3ef96c06f45707ad6 hoary-rc-install-i386.iso
bc72fe86a5b6e6ece80b0847c352549d hoary-rc-install-powerpc.iso
b94c4e79378b22282c40e20c0528ad9d hoary-rc-live-amd64.iso
21debcfa2d379063c0d5c2e6bf73dca7 hoary-rc-live-i386.iso
5e888b01b8652d12c6a7178f1f8947b1 hoary-rc-live-powerpc.iso

The md5sums for the 20050404 daily build are:

f7e5788e0fa6908756708af3c1f391e7 hoary-install-amd64.iso
d73a053a9c9a86180658acfc94923127 hoary-install-i386.iso
b49d3c69e638f35979d1880de8e84cde hoary-install-powerpc.iso
55beedf1329d313dfc631392ad2667ac hoary-live-amd64.iso
80da5e7eecf13251a5615ad1a00ca25b hoary-live-i386.iso
f451ad62b31521ade4709700505e453b hoary-live-powerpc.iso

The content on our side is all correct. If you still believe there is a
problem, please send mail to <email address hidden>; Bugzilla is not the
best place to discuss it.

(In reply to comment #100)
> (In reply to comment #99)
> > It fails to load ide-{mod,probe-mod,detect,floppy}.
>
> Those messages are confusing but normal, and the installer only displays them to
> the user in expert mode.
>

Regardless, it still fails to find or mount the CD, and therefore to install
ubuntu. When I go to VC2 and type:

mkdir /tmp/foo
mount -r /dev/cdroms/cdrom0 /tmp/foo
It says:
mount: Mounting /dev/cdroms/cdrom0 on /tmp/foo failed: Invalid argument

There's nothing new on VC4 when I do that. It also won't mount my thumb drive,
same error.

When I tell the installer screen on VC1 to try mounting it again, the error is:
Apr 4 19:40:52 cdrom-detect: CDROM-mount failed (error=1):
device=/dev/cdroms/cdrom0
It does that twice, with the usual messages about retrying, unmounting, and
giving up.

I tried mounting the device file that the cdrom0 link points to directly, same
error. Ditto for the thumb drive.

--jh--

(In reply to comment #101)
> (In reply to comment #100)
> > (In reply to comment #99)
> > > It fails to load ide-{mod,probe-mod,detect,floppy}.
> >
> > Those messages are confusing but normal, and the installer only displays them to
> > the user in expert mode.
> >
>
> Regardless, it still fails to find or mount the CD, and therefore to install
> ubuntu. When I go to VC2 and type:
>
> mkdir /tmp/foo
> mount -r /dev/cdroms/cdrom0 /tmp/foo
> It says:
> mount: Mounting /dev/cdroms/cdrom0 on /tmp/foo failed: Invalid argument
>
> There's nothing new on VC4 when I do that. It also won't mount my thumb drive,
> same error.
>
> When I tell the installer screen on VC1 to try mounting it again, the error is:
> Apr 4 19:40:52 cdrom-detect: CDROM-mount failed (error=1):
> device=/dev/cdroms/cdrom0
> It does that twice, with the usual messages about retrying, unmounting, and
> giving up.
>
> I tried mounting the device file that the cdrom0 link points to directly, same
> error. Ditto for the thumb drive.

If the device file appears, but doesn't work, then yours is a different bug.
Please open a new one.

Well, I've still this bug on a regular Hoary CD install.

Dell computer with two SATA HD and a NEC DVD+RW ND-3530-A

https://bugzilla.ubuntu.com/show_bug.cgi?id=8396
http://ubuntuforums.org/showthread.php?t=14464

I actually still suffer from this problem, not only with Ubuntu but also with Fedora(and Enterprise), SUSE, Debian, etc.

Information is loaded from the CD and then it can't detect both devices.

The only active system now is windows so can't do any analysis on the problem.

I tried all the solutions that are available on different forums(and the whole list here and they all don't work).

High Priority ?

This problem is a subset of 85578.
Apparently first identified 2 1/2 years ago and still not fixed Is anyone at Ubuntu really concerned?
Fedora, KNOPPIX and Mepis don't have this problem.

yeti

The original bug here is that ATAPI drives were not being detected by libata. That issue was fixed by enabling ATAPI in the libata header.

If people are still having issues, they are need to file separate bug reports. Most likely it is a case of the PATA/SATA driver not being present, so include lspci -vv and lspci -vvn output for your system.

--- Ben Collins <email address hidden> wrote:

> The original bug here is that ATAPI drives were not
> being detected by
> libata. That issue was fixed by enabling ATAPI in
> the libata header.
>
> If people are still having issues, they are need to
> file separate bug
> reports. Most likely it is a case of the PATA/SATA
> driver not being
> present, so include lspci -vv and lspci -vvn output
> for your system.
>
> --
> ATAPI CD-ROM not detected in system with SATA
> https://launchpad.net/bugs/8196

Ben,

Please read 85578 from the beginning.
Note the material I have already supplied.
Maybe the problem is in 'the installer'.
Anyways, _lots_ of Ubuntu people have this problem.
I suspect Fedora, Knoppix and I _know_ Mepis people
_don't_ have this problem. I'd like my Ubuntu to
work as well as my Mepis.

Respectfully,
yeti

yeti
>

____________________________________________________________________________________
Bored stiff? Loosen up...
Download and play hundreds of games for free on Yahoo! Games.
http://games.yahoo.com/games/front

yeti wrote:
> I suspect Fedora, Knoppix and I _know_ Mepis people
> _don't_ have this problem. I'd like my Ubuntu to
> work as well as my Mepis.
>
If Mepis detects these drives then it should be straightforward for us
to make sure we do the same.

--- Mark Shuttleworth <email address hidden> wrote:

> yeti wrote:
> > I suspect Fedora, Knoppix and I _know_ Mepis
 people
> > _don't_ have this problem. I'd like my Ubuntu to
> > work as well as my Mepis.
> >
> If Mepis detects these drives then it should be
> straightforward for us
> to make sure we do the same.
>
> --
> ATAPI CD-ROM not detected in system with SATA
> https://launchpad.net/bugs/8196
>

Thanks for your concern, Mark.

FYI, my version of the 8196 complaint, and some
supporting data are set forth in 85578.
My problem is probably peculiar to Gigabyte
motherboard and use of SATA primary with ATAPI
secondary.

Regards,
yeti

____________________________________________________________________________________
Have a burning question?
Go to www.Answers.yahoo.com and get answers from real people who know.

On Sat, Feb 24, 2007 at 06:26:31PM -0000, yeti wrote:
> FYI, my version of the 8196 complaint, and some supporting data are set
> forth in 85578. My problem is probably peculiar to Gigabyte motherboard
> and use of SATA primary with ATAPI secondary.

That bug is open and assigned to Ben for investigation. Please direct any
further comments there. As Ben explained, this bug is about an older
problem which has been fixed, and any similar issues are unrelated and need
to be addressed separately.

Thanks,

--
 - mdz