[sane-devel] a second try

Henning Meier-Geinitz henning at meier-geinitz.de
Tue Aug 2 17:01:51 UTC 2005


Hi,

On Mon, Aug 01, 2005 at 08:48:31PM +0200, Dieter Jurzitza wrote:
> Dear Henning,
> with pleasure :-)
> I wrote a short batch and echoed start / stop sequences into /var/log/messages 
> so the excerpt of /var/log/messages I have attached relates exactly to the 
> script I used to start scanimage.

There is no error in the SANE log so for me it looks like SANE works
correctly. This is the command that takes the longest time:

dev_max(currently)=32 max_active_device=5 (origin 1)
 def_reserved_size=32768
 >>> device=sg4 scsi1 chan=0 id=6 lun=0   em=0 sg_tablesize=128 excl=1
   FD(1): timeout=300000ms bufflen=131072 (res)sgat=4 low_dma=0
   cmd_q=1 f_packid=0 k_orphan=0 closed=0
     rb>> act: id=5 blen=48 t_o/elap=300000/7ms sgat=0 op=0x24
[sanei_scsi] scsi_req_enter: queue_used: 1, queue_max: 2
[sanei_scsi] sanei_scsi_req_wait: waiting for 0x4028d008
[sanei_scsi] sanei_scsi.issue: 0x4028d008
dev_max(currently)=32 max_active_device=5 (origin 1)
 def_reserved_size=32768
 >>> device=sg4 scsi1 chan=0 id=6 lun=0   em=0 sg_tablesize=128 excl=1
   FD(1): timeout=300000ms bufflen=131072 (res)sgat=4 low_dma=0
   cmd_q=1 f_packid=0 k_orphan=0 closed=0
     rb>> rcv: id=5 blen=48 dur=31106ms sgat=0 op=0x24

That means, the timeout was 300 seconds (manually set by you?) and the
time really needed was 31.1 seconds. This is the "set window" command
which seems to wake up the scanner. So there is no timeout here.

I don't know the reason for the timeout in your kernel log. Maybe the
SCSI controller doesn't expect such a long taking command? Anyway, no
error is send to the scsi generic level or to SANE.

You could try to comment out everything but "mustek" in dll.conf,
maybe this is cause by another backend?

Also I'm not sure which device generates the timeout. The SANE log says:
 >>> device=sg4 scsi1 chan=0 id=6 lun=0   em=0 sg_tablesize=128 excl=1
while the kernel log says:
Aug  1 20:40:41 djunix kernel: scsi1:0:5:0: Attempting to queue an ABORT message

So it looks like these are different devices (what is SCSI id 5 on
that bus?)

Bye,
  Henning



More information about the sane-devel mailing list