https://qa.mandrakesoft.com/show_bug.cgi?id=930
------- Additional Comments From [EMAIL PROTECTED] 2003-01-22 17:55 ------- "Andrey Borzenkov" <[EMAIL PROTECTED]> writes: Thierry, bug 930 is assigned you but it is pure kernel problem actually :) > > Have a LS120 removable drive. [...] [...] > << sigh >> wish i knew to look here earlier. > Here's a sip from the file: > > scsi0 : SCSI host adapter emulation for IDE ATAPI devices > Vendor: MITBISHI Model: LS-120 SLIM 05 Rev: 0015 > Type: Direct-Access ANSI SCSI revision: 00 > Vendor: MATSHITA Model: DVD-ROM SR-8171 Rev: 059d > Type: CD-ROM ANSI SCSI revision: 02 > Attached scsi removable disk sda at scsi0, channel 0, id 0, lun 0 > sda : READ CAPACITY failed. > sda : dUmMy= 0, message = 00, host = 0, driver = 28 > Current sd00:00: sense key Not Ready > Additional sense indicates Medium not present > sda : block size assumed to be 512 bytes, disk size 1GB. > /dev/scsi/host0/bus0/target0/lun0: I/O error: dev 08:00, sector 0 > I/O error: dev 08:00, sector 0 > I/O error: dev 08:00, sector 2097144 > I/O error: dev 08:00, sector 2097144 > I/O error: dev 08:00, sector 0 > I/O error: dev 08:00, sector 0 > ldm_validate_partition_table(): Disk read failed. So it confirms my guess, the problem is caused by sd driver together with a bug in \ ide-scsi in new kernel that makes ide-floppy devices appear as SCSI ... Here is the patch to sd.c that - sets size of media to 0 if no media is found (that stops kernel from attempting to \ read media to find partition type) - shuts up messages for this case (as it happens very often and automatically under \ devfs, messages are quite annoying). I'd like to ask you opinion if it should be so \ or it should still output a line about "Media not present" Here is the second patch for ppa.c that shuts up this cable warning that fills up my \ logs (and if you use supermount you have almost _only_ those cable warnings). It adds \ modules parameter no_cable_warning to supress it. Again the question is should I \ supress even more messages? I personally would like to do it as again, ppa is \ attempted to be loaded on many devfs lookups which is quite annoying. Please test them, if it OK they go to our kernel guys (not in this form). -andrey PS Please Cc to me on replies. I do read cooker but off the web so ... ------- 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] description: - plug a 100MB parallel zip drive on the parallel port - remove any zip from the drive - check with "cdrecord -scanbus" that the zip drive is actually installed (ppa module) - run: "find /dev/scsi" --> error ldm_validate_partition_table() : read disk failed The error appears on the device representing the ZIP drive. WARNING : this problem is causing a slow login : ********************************************************** At each first user login, the owner of the /dev entries are changed (devfsd?). This leads to an error when changing owner of the Zip device. Since the query and the error takes about 1,5 sec and they are repeated 3 times at login, the login is considerably slowed down, and takes up to 5 sec.
