M E M O R A N D U M 
  
  
  TO:     Pauline Judge
  
  FROM:   Keith Corbett
  
  DATE:   Aug. 19, 1986 
  
  SUBJ:   MCC ILLOP Crashes
  
  COPY:   John Salis
  
  
  MCC/CAD has taken a number of actions with their code in the past two
  weeks that may have solved some of their ILLOP problems.  Only one of
  these, reported by Jim Thomas, is related to a known, confirmed bug.
  The status of ILLOPs from Parallel Processing is not known because the
  principal users are not in this week.
  
  Accordingly, please put the ILLOP DCRs (list attached) on "monitor"
  status and/or update the log to reflect the indicated status.  I have
  closed one ILLOP call (#5129 from John Hardin) since the problem has not
  recurred since the suggested work-around was applied.
  
  We will stop monitoring software calls when the customer confirms that
  they have not seen the problem for over a week.  Meanwhile they should
  be removed from the daily open software report and should appear on the
  monitored call report.  Please copy me on monitored software call
  reports from now on.
  
  Please continue to open a DCR for any ILLOPs that are reported, and
  update the problem history for the open ones when customers call with
  subsequent crash reports pertaining to the same ILLOP number.


  KMC


DCR #   CALLER  ILLOP   STATUS
-----   ------  -----   ------

5478    K.Hammer  ?     She is investigating their code for possible
                        violation of memory conventions; will call back.
                        STAYS OPEN.

5446    J.Thomas 31345  We cannot reproduce the exact ILLOP, but seems
                        related to crash caused by (^ 3000 3000), which
                        is being investigated by Robert Putnam. 
                        MONITOR.

5368    J.Hardin 27037  Possibly caused by MCC code; he is not using his
                        Lambda anymore, but others have seen it.
                        MONITOR.

5341    McGarity 27037  He's out of town this week; left messages.  No
                        way to tell if he was still experiencing this
                        crash until he comes back; see #5528.
                        STAYS OPEN.

5499    VonBlucher 31336 Was assigned to H/W, but could be S/W.
                        I asked Jeff to call me if problem recurs, and
                        pass on the info on WITH-OUTPUT-TO-STRING which
                        caused this same ILLOP in CAD.
                        MONITOR.

5528    Calhoun  27037  Possible H/W problem, but monitor to see if it
                        recurs after reformatting disk.  If it does,
                        Miran will call 800# with an update to this DCR;
                        problem will have to be investigated between
                        Engineering and Pat McGarity, who wrote the code
                        Miran is using.
                        MONITOR.