https://bugzilla.novell.com/show_bug.cgi?id=326952#c39
Tejun Heo
(In reply to comment #37 from Tejun Heo)
Slave-only configuration reported by James doesn't have anything to do with what Ralf is reporting.
But comment #26 states: "I set the top drive (Pioneer) to Master via its jumper, and the bottom (Sony) to Slave via its jumper, removed the "T", and 10.3 found both drives."
libata issues SRST during probing while IDE doesn't. The broken configuration fails SRST protocol. However, doing SRST is more robust and correct for most cases. It's broken hardware configuration which just happened to work with previous configuration by dumb luck. Can't really modify driver behavior to support that particular broken configuration.
And no, you don't need to add retrying logic at HAL. I was just curious why restarting haldaemon makes any difference.
I expect a second try, or just a try some time later, just works.
I see.
Another thing is that the drive is also timing out commands w/o any data transfer - TEST UNIT READY, which is quite weird. There's a way to disable media presence polling, right? Can you please tell James how to disable it? I wanna know whether the mount failure is a result of polling failure or indicates something else.
To disable polling just do: $ hal-disable-polling --dev=/dev/sr0 Polling for drive /dev/sr0 have been disabled. The fdi file written was /etc/hal/fdi/information/media-check-disable-storage_model_DVD_RAM_UJ_842.fdi
Ralf, please disable polling using the above command and see whether mount still fails. Thanks. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.