Bugzilla – Bug 144861
cdrdao causes SCSI bus reset
Last modified: 2006-01-31 08:56:27 UTC
every time I run cdrdao scanbus it hangs for a while and I get this in /var/log/messages: Jan 23 17:43:28 g241 klogd: (scsi0:A:6:0): No or incomplete CDB sent to device. Jan 23 17:43:28 g241 klogd: scsi0: Issued Channel A Bus Reset. 1 SCBs aborted As root it prints # cdrdao scanbus Cdrdao version 1.2.0 - (C) Andreas Mueller <andreas@daneb.de> SCSI interface library - (C) Joerg Schilling Paranoia DAE library - (C) Monty Check http://cdrdao.sourceforge.net/drives.html#dt for current driver tables. Using libscg version 'schily-0.8' 0,6,0 : PIONEER , DVD-ROM DVD-303 , 1.10 # As user it doesn't. The reset message happens as user also. I suppose cdrdao does something nasty to the device.
Created attachment 64570 [details] hwinfo --scsi
Linux g241 2.6.15-git12-6-default #1 Tue Jan 17 14:22:14 UTC 2006 i686 i686 i386 GNU/Linux
The message is specific to the aic7xxx driver. Does it happen with cdrecord -scanbus ? The scanbus code in cdrecord is quite similar. Can you please attach output of cdrecord -scanbus -V ? Or better, can I get access to this machine?
it's g241, no NIS, std root pw. hald polling also causes problems (#145147), therefore it's not running atm. The machine is not running all the time because it's quite loud, please tell me if you're done so I can switch it off again :-)
I could not reproduce the messages, they were probably caused by hald too. The hang is caused by ATA devices detection. If it finds none, it falls back to obsolete interface which causes the delay. It could be eventually disabled. As user, resmgr does not report the scsi device, so it is not scanned. ATA devices are scanned anyway.
Ok, thanks. I'll have to have a look at the resmgr scsi thing in cdrdao then. resmgr doesn't report scsi devices on purpose anymore as the sg device can now get an acl as well.
Scanning the old ATAPI interface is now disabled. The delay should no longer appear.