So, with a little but of experimentation, I was able to get things working. Using the command SANE_DEBUG_UMAX=255 scanimage
I observed the following during the initialization(only the four lines that were relevant to me are shown): [umax] attach_scanner: sanei_scsi_open_extended returned scsi buffer size = 16384 [umax] variable scsi buffer size (usage of sanei_scsi_open_extended) [umax] sane_start: umax_scsi_open_extended returned scsi buffer size = 34000 [umax] bytes per line = 2550 and then the following during reading: [umax] umax_queue_read_image_data_req for buffer[0], length = 33150 [umax] umax_queue_read_image_data_req: id for buffer[0] is 0x8066630 [umax] umax_reader_process: read image data queued for buffer[0] [umax] umax_wait_queued_image_data for buffer[1] (id=0x8077088) [umax] sane_read: read 32768 bytes [umax] umax_wait_queued_image_data: wait returned status Error during device I/O [umax] ERROR: umax_reader_process: unable to get image data from scanner! [umax] reader_process: freeing SCSI buffer[1] [umax] reader_process: finished reading data [umax] sane_read: read 382 bytes [umax] sane_read: read 0 bytes [umax] do_cancel [umax] killing reader_process Since the read seemed to fail after 32768 bytes, I changed the /etc/sane.d/umax.conf file as follows: $diff umax.conf_save umax.conf 13c13 < #option scan-lines 100 --- > option scan-lines 12 So that the read would ask for less than 32768 (12*2550) and thus complete. Once I do that, all my scans work fine. So my question is why do I need to do this? Is this a bug in the umax backend? Or something else? Thanks. Cliff Romash
