RFC 477 (rfc477) - Page 2 of 19


Remote Job Service at UCSB



Alternative Format: Original Text Document



RFC 477                Remote Job Service at UCSB            23 May 1973


Service Provided

   The UCSB Computer Center operates an IBM 360/75 and runs OS MVT
   release 21.0 and HASP.  All batch jobs at UCSB enter the system
   through HASP, and therefore RJS acts as an interface between the user
   and HASP.  RJS's function is to provide the user with a HASP virtual
   RJE terminal, consisting of card reader, card punch, line printer and
   operator's console, and to manage the first three of these for the
   user in response to commands issued by him.  By virtue of the fact
   that RJS maintains the correspondence between a particular user and
   the HASP RJE terminal owned by him, the user may issue commands to
   alter the status of those jobs submitted by him through his RJE
   terminal.  This may be done even if the user has logged out of RJS
   and back in again, possibly from another site.

   The sequence of events involved in using RJS are as follows.  First,
   the user logs in, specifying a user name, password, and account
   number.  In addition to indicating how subsequent use of RJS is to be
   billed, this accounting information identifies the owner of a
   particular RJE terminal.  That is, the association between user name
   and HASP virtual RJE terminal is unique, and only one individual is
   allowed logged in under a given user name at a time.

   At present, billing within RJS is not implemented, and therefore the
   login account number is completely arbitrary, and user name and
   password are relatively so.  The first time a given user name is
   used, any password may be chosen; thereafter, as long as the user
   name is known to RJS, the same password must be used.  A user name
   remains known to RJS while he is logged in, and when he is not, as
   long as he owns at least one job known to PJS.  Otherwise, the user
   name is discarded.

   After a user is logged in, he specifies input and output file
   information and requests input retrieval be initiated.  He may then
   log out or not as he wishes; specifically he is not required to
   remain logged in during source file retrieval.  A job can completed
   abnormally at any stage of processing, or normally, after storage of
   the last of its output.  In any case, the circumstances of the final
   disposition of the job are displayed to the user immediately if he is
   logged in, saved for a period after its completion, and are available
   to him on request the next time he logs in.  This status information
   is retained for at least a fixed period (currently two days), but
   will be retained longer as long as sufficient storage space is
   available for it.







Krilanovich