Distribution: Debian testing Hardware Environment: Vaio PCG-C1MHP + iPod 30 gig (2003) Software Environment: gtkpod Problem Description: I had the same problem as described under bug #1872 with 2.6.1. Now I can connect to iPod with 2.6.3 now, but the connection brakes after a short time and the ipod is reconnected at sdc while it was on sdb before and so on, what leafs my gtkpod app waiting forever, since I can't manage to umount the lost sdb2 connection to remount sdc2 to /mnt/ipod. I have compiled ieee1394, ohci1394, sbp2 and sd_mod directly into kernel. If you need the kernel config or something else, just ask. By the way all this was working in 2.4.20+ but getting the device connected was far more work, so thanks for that part of your work! I have added my dmesg log. Steps to reproduce: Take iPod, mount somewhere transfer big amount of data (I think I have to write to iPod, since I could download several gigs without problems) or wait a while and voila, there it is.
Created attachment 2306 [details] dmesg output log dmesg.log
Does this still occur with 2.6.11.12 or 2.6.12-rc6? Thanks, Nish
The repeated bus resets that are shown in your dmesg log indicate that there is a harwdare problem: Bad cable, noisy controller, noisy front panel or back panel connector... Linux 2.4 simply handles such conditions a bit more graceful than the 2.6 version you have installed. You could try 2.6.12 which may a bit more successful with such problems as well, although there are still regressions from 2.4 to 2.6.12.
I looked up information about Vaio PCG-C1MHP Picturebook. I assume you use the onboard connector of the Picturebook, i.e. there are no such things as front or back panel connectors which could introduce noise. A bad cable, EMI or heat problems in the Picturebook or even in the iPod remain as possible causes for the repeated bus resets. The problem that scsi disk device files are enumerated through sda, sdb, etc. were fixed at some point in Linux 2.6. I don't remember the particular subrelease. However, these problems may be introduced again at any new Linux 2.6 release since we still struggle to get hotplugging right between sbp2 and the scsi subsystem. However, I changed this bug to "rejected/ will_not_fix" because we cannot fix the underlying hardware problem. We can only try to make the drivers as robust as possible in situations of bus resets. (And I hope we will get Linux 2.6 at least as robust as 2.4 eventually.) But ultimately we cannot provide a software solution to an evident hardware fault.
The hot(un)plug trouble between sbp2 and SCSI subsystem that I referred to in the previous comment has been fixed in 2.6.14-rc3.