[H-GEN] Yamaha 4416 CDRW and cdrecord

[Paul Gearon]

[ Humbug *General* list - semi-serious discussions about Humbug and  ]
[ Unix-related topics.  Please observe the list's charter.           ]
[ Worthwhile understanding: http://www.humbug.org.au/netiquette.html ]

Hi all,

I've a bit of a problem with my CDRW, and I'm hoping someone out there may
have heard of it.  It's a kind of longish email, so if you don't know
anything here then please don't feel obliged to read on.

I have a Yamaha 4416 SCSI drive which used to work quite well on my old
machine (which had an Adaptec 2940 SCSI-2 bus).  Then I upgraded to a
system with an Ultra 2 bus and things started to go downhill.

First, a little history:

On a 2.2.x kernel I was OK, but once I went to 2.3.x (and now 2.4.x) I
found that the scan of the scsi chain would take over 5 minutes (it should
only be a few seconds).  I put the drive in an external case, and I found
that just having power to the drive was enough to cause this problem.
Booting up with the drive unpowered, and then switching it on and adding
it to the kernel's chain with an 'echo "scsi add-single-device 1 0 1 0" >
/proc/scsi/scsi' would also take several minutes.

Fortunately, through all of this, the drive worked a treat.  Since I
reboot rarely (I shouldn't need to at all, but until recently I couldn't
play DVDs under Linux) the long bus scans didn't bother me too much, but
they were annoying.

Since the old kernel didn't have this problem I thought it was possibly a
kernel bug.  Just recently though I went looking for the latest firmware
on the 4416 and discovered that there was a bug in the drive which
prevented it from being recognised when attached to the bottom half of a
U2 or UW SCSI bus (ahhh, so that's why the drive never showed up in
Windows.  I have no Windows cdr software so it never occurred to me to get
it working in Windows).  So I went and updated the firmware on the drive
(the firmware update was a Windows program, so I used a different PC which
had a narrow bus and Win98).  The firmware claimed it was only tested with
Win98, but of course, that's not unusual. After re-attaching the drive to
my own PC with the U2 bus, the machine now boots up quickly, with the SCSI
bus scans going perfectly.  (Yay!)

Now for the problem:

Since the firmware update I can no longer reliably burn a CD.  I've had
some success, but generally it's been for ISOs which only contain 1
enourmous file.  Anything else and it dies.

For instance, I tried burning the latest Debian ISOs.  The first ISO would
get almost the entire way through the burn when cdrecord would report a
"Sense Key" error with a "Sense Bytes" line giving heap of hex digits
which I don't follow (sorry, I don't have the exact text here, but I can
get it if someone wants it).  The error is reported as "Retryable" and the
burn continues.  Some of the resulting CD is readable, but several files
don't work.  The second iso only gets a few MB into the burn when it
reports the same erorr (again, Retryable) and finishes the burn only a
couple of MB later.  The resulting CD of is of course a mess.

The interesting thing is that for ISO files which fail, then these errors
occur at EXACTLY the same place in the file during each burn (I've wasted
several CDRs demonstrating this).  So the first Debian ISO consistently
fails neat the end of the burn, and the second Debian ISO consistently
fails after only 8MB of the burn.  I have no idea what could be in a file
to cause this (I can't imagine how the data in a file could have any
effect on the process), and as I've already mentioned, I've built a couple
(not many) of ISOs which can burn just fine.

Same sort of deal for music CDs.  I burnt a CD recently that was fine.
Then the other day I went to burn another music CD which got all the way
to track 10 before reporting the "retryable" error (ironically, this is
music recorded by my musician brother-in-law, so I'm not violating
copyright here!).  The resulting CD cannot be read by either a CDROM or a
CD player (both spin up, and then stop after they fail reading the index,
I guess).

I was tempted to just downgrade the firmware, but of course, Yamaha don't
offer old firmwares for their drives.  :-(  At this point I'm wondering if
I'll need to sell the drive to a Windows 98 user (which would be tough,
since most of them won't have a SCSI card) and to buy myself a new drive.
It _should_ work for a Windows user, but I don't have any software to test
that theory.

In the meantime, I've written to Yamaha (not that I expect a response for
a Linux problem) and am thinking I should get the exact output of cdrecord
and email the author.

So has anyone heard of anything like this?  Any suggestions?  I'm kinda
desperate as I NEED to back stuff up, and I can't afford a tape drive.

Regards,
Paul Gearon

Software Engineer                Telephone:   +61 7 2876 2188
Plugged In Software              Fax:         +61 7 3876 4899
http://www.PIsoftware.com        PGP Key available via finger

Catapultam habeo. Nisi pecuniam omnem mihi dabis, ad caput tuum saxum
immane mittam.
(Translation from latin: "I have a catapult. Give me all the money,
or I will fling an enormous rock at your head.")



--
* This is list (humbug) general handled by majordomo at lists.humbug.org.au .
* Postings to this list are only accepted from subscribed addresses of
* lists 'general' or 'general-post'.