http://qa.mandrakesoft.com/show_bug.cgi?id=5692
------- Additional Comments From [EMAIL PROTECTED] 2003-15-09 18:01 ------- Duane, latest supermount is supposed to allow you to get the disk from the drive, even when files are still "open". If not, it is either a bug or you have disabled this in fstab. Can you give a clear test-case (+fstab file)? d. -- Configure bugmail: http://qa.mandrakesoft.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is. ------- Reminder: ------- assigned_to: [EMAIL PROTECTED] status: UNCONFIRMED creation_date: description: This is probably more of a design issue than a bug but I want to hear the reasons for, and encourage a re-thinking of, the current design. It was really silly for CD drive manufacturers to add a software disable for the eject button on the front of a READ-ONLY device. I blame them initially for all this frustration/confusion, but we programmers could fix this problem if we want. The general problem: more and more things that used to be simple about computers are becoming complex. Enter this particular example: often the cdrom won't give you the disk and there is usually "no good reason" for it to refuse. I'm an OS programmer. I know handling I/O faults is a pain and have often wanted to ignore the error paths and events that abort device reads (especially the async reads). But locking the drive door for a read-only device just because we can is a really bad idea. This makes for an UNresponsive machine that does not act the way it is expected and ultimately creates distrust in the user about his system and our software. I should NOT have to type 'eject' to make the drive open. (This is not a MAC, there IS a button, so it needs to work whenever possible) The errors must be handled anyway so the OS and all the apps that use the device must always be prepared for "media is gone". I don't care if the device is mounted or even in the middle of a read, when I push the open door button I expect to hear the device spin down immediately and open the door - no questions asked. The kernel and apps should recover from these errors gracefully - it should really be no big deal. I strongly encourage a rethinking of this design. Fight the urge to lock removable media drives unless *and only while* they are writing. We want machines that are responsive. Beeping the PC speaker would be a good way to say the software knows the user pressed the open door button while the device is writing (but there probably isn't hardware support for this). An attitude like this will greatly improve the "look and feel" of the user's machine and our software. Copy this to any application writers that need to get involved. Next time I'll go after the Power Button... (but not on a software forum :)
