On Fri, Jul 25, 2003 at 12:32:50PM -0400, Ashwin Bijur wrote: > > >I am experiencing a problem with chg-scsi. Whenever I execute > >"chg-scsi -inventory", I get back the following response: > > BarCode == 1 and emubarcode == 1
This is saying you are claiming to have a barcode reader and asking chg-scsi to emulate one also. Only one or the other can be set to 1. > >Our tape-changer is equipped with a bar code reader. I have set > >"havebarcode 1", "emubarcode 0" in the chg-scsi.conf file. But > >chg-scsi doesn't seem to pickup this information. When I ran "strings > >-a /usr/local/etc/libexec/chg-scsi | grep emubarcode", I got the > >following output: > > > >emubarcode > >emubarcode 1 # If you drive has no barcode reader this will try > >BarCode == 1 and emubarcode == 1 > >emubarcode > > > >So it seems that emubarcode is hardcoded as 1 in the binary file. Is > >there any way to fix this or force chg-scsi to read the value for > >emubarcode from the chg-scsi.conf file? There is nothing in that grep output to suggest emubarcode is hardcoded as 1; == is a test, not an assignment. > >When I run "chg-scsi -genconf", I get the following information: ... > >emubarcode 1 # If you drive has no barcode reader this will try > > # keep an inventory of your tapes to find them faster > > # > >havebarcode 0 # Set this to 1 if you have an library with an I peeked at the chg-scsi.c source code. emubarcode is initialized to 0. However, I did note an oddity in the part of the code that reads the config file. Where is detects "havebarcode", it uses the value, 0 or 1, that follows the configuration parameter. This is as expected. When "emubarcode" is detected it is treated differently. It is treated as a boolean and the value that follows is ignored. I.e. if "emubarcode" is in chg-scsi.conf, THEN it is set to 1 regardless of the value. I suggest you try eliminating the emubarcode entirely from your chg-scsi.conf file (still setting "havebarcode 1") and see if that works. It should pick-up the hardcoded emubarcode default of 0. If that works, we can claim to the maintainers of the code that it is a defect in the documentation or the source code. -- Jon H. LaBadie [EMAIL PROTECTED] JG Computing 4455 Province Line Road (609) 252-0159 Princeton, NJ 08540-4322 (609) 683-7220 (fax)
