On Sat, 18 Dec 2010, [ISO-8859-15] Richard Schütz wrote: > I experience a problem when I transmit payload and issue some ATA > commands (e.g. querying SMART or using hdparm) at the same time to an > external hard drive which is connected via USB. > > > Kernel: 2.6.36.2 > USB Controller: Intel Corporation 82801I (ICH9 Family) USB2 EHCI > Controller #1 (rev 03) ... > Without simultaneous data transmission querying SMART works fine: ... > Now let's transmit some data: > > #dd if=/dev/sdb1 of=/dev/null > > > And now query SMART again: > > The "INFORMATION SECTION" is still retrieved quickly, but then it hangs > after printing "=== START OF READ SMART DATA SECTION ===" and the status > LED of the bridge stops flashing and after a short delay there are some > errors in the kernel output: > > Dec 18 17:09:23 myhost kernel: usb 2-2: reset high speed USB device > using ehci_hcd and address 4 > Dec 18 17:09:24 myhost kernel: sd 7:0:0:0: [sdb] Unhandled sense code > Dec 18 17:09:24 myhost kernel: sd 7:0:0:0: [sdb] Result: hostbyte=0x00 > driverbyte=0x08 > Dec 18 17:09:24 myhost kernel: sd 7:0:0:0: [sdb] Sense Key : 0x3 [current] > Dec 18 17:09:24 myhost kernel: sd 7:0:0:0: [sdb] ASC=0x11 ASCQ=0x0 > Dec 18 17:09:24 myhost kernel: sd 7:0:0:0: [sdb] CDB: cdb[0]=0x28: 28 00 > 00 1b 6b 9f 00 00 f0 00 > Dec 18 17:09:24 myhost kernel: end_request: I/O error, dev sdb, sector > 1797023 > Dec 18 17:09:24 myhost kernel: Buffer I/O error on device sdb1, logical > block 224620 Sense key 3 and ASC 0x11 means Unrecovered Read Error, for whatever good that does. > NOTE: This is reproducible with another USB ATA bridge which supports > SAT (smartctl -d sat), too! > > I'm not sure if this is a common problem of this SATA USB bridges (as > it's reproducible with my other and different external drive, too) or a > problem with the USB stack and/or the usb-storage driver handling this > situation which needs to be fixed. Most probably it is not a problem in the kernel, meaning it _is_ a bug in the SATA-USB bridge or the drive. You can get more information by recording a usbmon trace of a failure; see the kernel source file Documentation/usb/usbmon.txt. Alan Stern -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
