On 01/12/05, Randall R Schulz <rschulz@sonic.net> wrote:
Hi,
Recently I began to have trouble with YaST refusing to use the SuSE 10.0 commercial DVD for new package installations. It claimed there wasn't a distribution on the device, though clearly there was, as I'd installed from it many times already. Examining it via the shell seemed to show everything being in order, but when I went to look at .../media.1/media, I got this:
% cat /media/SU1000_001/media.1/media cat: /media/SU1000_001/media.1/media: Input/output error
Since I have two DVD recorders in this system, I switched to the other one and when using that drive, everything's working fine. YaST will install from that drive and cat shows the expected contents of the .../media.1/media file.
Experimenting further, I tried these commands:
% dd if=/dev/hda of=/dev/null bs=4096 1048575+1 records in 1048575+1 records out 4294965248 bytes (4.3 GB) copied, 534.002 seconds, 8.0 MB/s
% dd if=/dev/hdc of=/dev/null bs=4096 2082936+0 records in 2082936+0 records out 8531705856 bytes (8.5 GB) copied, 1089.81 seconds, 7.8 MB/s
/dev/hda is the misbehaving drive and /dev/hdc is the one that works.
Hda is a Sony DRU-720A dual-layer writer, hdc is an older Sony DRU-510A single-layer writer (both are capable of reading dual-layer media). Both are IDE drives configured as masters of the two separate IDE buses of my mainboard. There are no slaves on those buses and all other mass storage is on SCSI.
It seems very odd that while no error was reported, reading stopped on hda after 4.3 GB while hdc when the full 8.5 GB. One might suppose that hda couldn't handle the layer switch yet reported no error when it should have done so.
Another oddity I noticed when conducting these experiments was high CPU usage when accessing hda. It turns out its DMA had gotten turned off somehow. I reenabled it and ran the test again. Aside from going about twice as fast, nothing changed.
Has anybody seen symptoms like this? Can you shed any light on what might have happened to the malfunctioning drive to account for these peculiar symptoms?
Thanks.
Randall Schulz
-- Personally, my suspicions would be with the laser lens. Maybe its calibration has gone awry or something? That's purely off the top of my head...
-- ============================================== I am only human, please forgive me if I make a mistake it is not deliberate. ============================================== Take care. Kevan Farmer Linux user #373362 34 Hill Street Cheslyn Hay Staffordshire WS6 7HR