lazlow, I agree.

I have tried replaceing the IDE cable, but no change. Also re-flashed the bios. Was at version 67, now up to version 87. Also didn't make a difference.

Thank you everyone that has made suggestions, I do appreciate all the responses and it has been helpful. The next step is to get a SATA DVDRW and just bypass the issue. Not real high on my priority list, but someday. These IDE drives would work well in an external USB enclosure or possibly a good upgrade for the CDRom in my oldest grandson's computer.

:D

I've had issues like this after getting a bad burn. A reboot normally puts the drive back into a normal state. Although there's plenty of dmesg entries I/O errors when it happens.

Try to put the drives on their own cables. Even if that means removing a drive. Make sure you have the right power connector going to it. More of an issue on older machines / cheap power supplies and the like. It SHOULD open when doing the initial boot process. CMOS / BIOS screen and probably even in GRUB. If not, then it's NOT a linux issue, but likely a hardware issue. It sounds like it's working for you to some degree, so probably good hardware (to a limited extent). Have you tried just eject -r (no other parms). Which drive does it assume for the default one?

More tests for DVD drives:

1 Press and hold down tray button for several seconds just after system powers on. If tray opens repeat step to see if problem is fixed. If tray(s) fails to open try step 2.

2 Press tray button repeatedly just as the system powers on. If tray opens repeat step to see if problem is fixed but press button only once. If tray(s) fails to open try step 3.

3 Disconnect IDE cable to DVD drive(s). Push tray button just after system powers on. If tray fails to open move drive to another PC and test again. If trays still won’t open drive(s) may be defective. If tray opens then try a new power supply in problem PC with IDE cable disconnected and test again. Connect IDE cable and test again. If tray(s) won’t open then BIOS is incompatible with DVD drives.

You took the words right out of my mouth.

I once had a *newer* CD burner that I installed in a PII system running Win98 that acted the same. Holding the button until the tray opened was the only way to open it when it was empty, whether or not the OS was booted. After upgrading and installing the same drive in a PIII with the same OS, it worked normal.

Having the IDE cables disconnected is the perfect test if step 1 or 2 doesn't work.

Cheers

Hi

Reviewing the problem:
intel D946GZ motherboard - single ATA channel

Tried alternate OS - Knoppix - same problem - check
Verified drives are set as master and slave - check
Replaced ATA cable (ensuring it is high density type) - check
Checked both drives are correctly identified by BIOS - check
Cannot eject either drive tray by pressing drive eject buttons on system power up - Hmmm - indicates possible hardware problem
Tried single DVD configured as master - check - Didn't work

Hmmm
Tried a single CDROM as master on IDE as suggested ?

More information requested:
(a) Is your mother board a D946GZIS, D946GZTS, or D946GZAB ?
(according to the intel www site there are three(3) in the D946GZ series)
(b) Would you please post the BIOS version number here.

Questions:
(a) Is it possible that the IDE interface has been explicitly disabled at the BIOS level ?
(b) Is it possible that the IDE interface has been implicitly disabled when you have selected to enable another feature on the motherboard ?
(c) Have you tried installing a single CRDOM as master on the IDE channel to see if it is the DVD that is an issue ?
(d) Have you tried attaching power to the DVD drive, without a data cable connected, and then tested the response to pressing the eject button ? (this will verify that the drive itself is ok)
(e) Have you ever flashed the BIOS on this motherboard ?
(perhaps the BIOS copy is flawed or the wrong version for your motherboard)
(f) Have you tried installing a single IDE PATA HDD on the IDE channel as a test ? (if a IDE PATA hard disk works, but the DVD doesn't that give us some more information to work with).
(g) Do you have both power supply cables (24pin & 4pin)connected to the motherboard ? In particular the 4 pin cable. This should make no difference but it is worth checking any way.
(h) With one DVD installed, and nominated as a boot device) does the drive front LED flash on system boot indicating that the drive is being accessed ?
(i) If you have a multimeter available check that the voltages are present on the power supply fly lead that you are connecting to the DVD drives. Perhaps your power supply is not generating the correct voltages (or is unable to supply sufficient current) to allow the DVD drives to operate. (ie. 5V or 12V lines are marginal, etc).
Also check that you are using a P4 (Or better) power supply (ie with a 4 pin (yellow and black) motherboard supply lead coming directly out of the power supply case; as opposed to using a lower rated power supply with a drive power connector to 4 pin motherboard power adapter cable).

Comment:
I'm sorry to repeat things, but like you I am suspicious that there may be a simple explanation to this problem. It sounds more like a hardware problem than a software problem.

Best regards

Chris

To cgtueno:

yes, have tried single CDROM (DVD in this case) set as master on the IDE cable. Have tried both drives that way, and tried with a different cable.

Information requested:
The motherboard is D946GZIS. I'm not at home so can't get the whole BIOS version number, but it was a level 67, and I reflashed it to level 87, which was the latest and greatest about a week ago.

I don't think the IDE is disabled, at least the bios sees them and recognizes both drives correctly. If there is a boot disk in either drive, the system can boot from them. I'm not sure if it can from the external USB drive as I don't think I've ever tried that, but would assume that it can. On boot the system does strobe both internal drives. I can borrow an IDE hdd from the techs, but haven't tried that yet.

The power supply is correctly hooked up, but I haven't actually checked the voltage. When I purchased this computer, I had them upgrade the power supply when it was built.

More things I'm going to try:

I just purchased a very similar DVDRW to the one that I have, but it's SATA instead of IDE. Also picked up a USB to IDE/SATA adapter cable and am going to try the drives on that.

With school coming to an end here (I support the core computers in a medium sized school district) I haven't had much time to mess with this lately so it may take a couple of days before I can try any more of the suggestions.

I will post what I find out because even if I don't resolve this, maybe someday one of the suggestions here might help someone else resolve a similar issue.

Thank you again to everyone for your hints, comments, and suggestions. This site always amazes me with how helpful everyone is.

Have you tryied mounting the drive first? I think the issue here is with HAL, unless it is mounted you can't really do anything. This perticular issue seems to pop up too often.

Mounted or not, if there's a disk in the drive, it works like normal. The failure is only when there isn't a disk in the drive. Most of the time I leave the drives with a non-bootable disk in them and they work like you would expect from either the command line or by pushing the button on the front. When it fails is when I forget to leave a disk in the drive. Then I have to use a paperclip to open the drive. From that point on, they work fine. In fact, it's rare that either of these drives make a coaster when I'm burning CDs or DVDs.

Home now and have added a SATA DVDRW (LITE-ON iHAS324 A). Works perfectly.

Plugged the DVD/CDRW (DVD 0) into a USB to IDE converter cable and tried to eject and it failed. Still works for everything else, but still will not eject empty. Tried plugging into a different computer, and still won't eject empty. Un-plugged from the computer with just power plugged in to it, and it still will not eject.

I haven't tried the other IDE DVD drive yet, but I bet it'll act the same way. I think these drives both have issues. The two IDE drives are different, but they are approximately the same age.

Hi

Hmmm now that narrows it down a bit.
Same fault in a different machine.

I'm loathed to suggest it, but you could consider updating the firmware on the DVD drives themselves. That requires a loader application, and a copy of the updated firmware.

For example:
LH-20A1H
http://us.liteonit.com/us/index.php?...=204&Itemid=67

The reason I am loathed to suggest it, lies with the fact that the drives should operate normally with their original factory flashed firmware. You generally don't re-flash the firmware unless there is a bug or a feature enhancement (ie. new media types supported,etc).
A failure in the flash process can convert a usable drive to a useless brick if it goes wrong.

Have these drives suffered heavy wear and tear ?
Perhaps they are mechanically defective.
Have you carefully removed the front plastic bezel on each drive and pressed the eject button directly ? Is it possible that the bezel's plastic button is no longer making contact with the actual PCB drive eject button, or the buttons are damaged, or the wrong bezel has been fitted to the front of the drive ? (bezel = plastic front face plate).

When you test the eject button, make sure that you are testing it with the power connected only (not the data), then you are testing the drive stand alone, the eject button should work in this case.

Don't laugh but I have been given DVD drives in the past as scrap, that had the wrong bezels installed, and therefore the eject buttons didn't work. (Because the plastic bezel button was nowhere near the actual button on the PCB).

Not sure what to suggest after that.
Perhaps the drives are due for retirement to Silicon Heaven ?

Hope that helps

Chris

Hi

Just re-read the above again.
Strike that about the front bezels, since you state that the drives eject normally if there is media in the drive.

Hmmm

Chris

Perhaps there's something funny with your dbus + hal + udev setup.

See if the contents of /dev/ changes when going from the non-media non-working setup to the media and working setup.

$ ls -l /dev/* | tee ls_of_dev_bad.log
$ ls -l /dev/* | tee ls_of_dev_good.log

$ diff ls_of_dev_bad.log ls_of_dev_good.log

It might hint towards what's different / changing between setups. And therefor how to manage a possible solution. Not that it will, but it might. It could still be just firmware, or any number of combinations of things.

That's entirely possible. At least we have eliminated any possibility of a software problem so that’s progress.

Did you try any of the suggestions I made in post #18? If so what were the results?

To win_to_lin_migrant:

Yes, did try those things. None of them worked. Even tried holding the button on the drive and then plugging in the power cable that came with the USB to IDE converter cable - drive not even in a computer - no data connected at all.

To cgtueno:

I haven't done it yet, but I think I'll try flashing the firmware on one of them. I'll have to do it at work so it may not be right away.

To Shadow_7:

I have the very same problem when the drive is outside of the computer with only power connected - no data cable.


These drives have had moderate usage. I burn CDs and/or DVDs of pictures we take at the grandsons football games for each of the players on the teams. Between the two drives we probably have burned a few hundred disks.

I am thinking the problem is with the drives. The second computer that I tried this in is same brand, but much older - 500mhz ancient junk. And with it also having the issue with the external power supply kind of eliminates the computer / OS from the equation. The next step is to reflash the firmware. Just wish they had a linux or dos version of the flash program - I don't have a windows computer to put this in myself.

Thanks again everyone.

What I am about to suggest shouldn't work -- I, too, think it's a h/w problem in the optical drives.

But what about trying eject w/ the "-n", "-p", "-m", "-r", "-T", "-s", & even "-t" options?

I'd be especially interested to see the optical device is still being found by the OS when it is unresponsive, as well as if using /proc/mounts instead /etc/mtab makes any difference: