MS. LINDA BLIDEN
SYS-ED
ONE PARK AVENUE
NEW YORK N.Y. 10016

DEAR LINDA,

I WANTED TO LET YOU KNOW HOW MY PROBLEM WITH THE ALTERNATE INDEX AND OPEN ERRORS WORKED OUT IN CASE SOMEONE ELSE CAN USE THE INFORMATION.

AS I DESCRIBED TO YOU, WE WOULD GET AN OPEN ERROR ON ANY BASE CLUSTER THAT HAD AN ALTERNATE INDEX AND PATH DEFINED - IF THE PATH WAS DEFINED TO THE FCT. IN ALL HONESTY, THE SOLUTION I FINALLY CAME UP WITH IS A COMPROMISE. ALTHOUGH IT WORKS IT ISN'T REALLY CLEAN. I REASSEMBLED THE FCT SO THAT THE BASE CLUSTER IS OPEN DEFERRED AND THE PATH IS OPEN INITIAL. NO OPEN ERRORS WILL BE DETECTED WHEN THE ONLINE FIRST COMES UP. BUT, WHEN YOU OPEN THE BASE LIME (BY USE OF THE CICS 'CSMT' COMMAND IT GETS AN OPEN ERROR (AB '168'). TO REMEDY THIS, YOU MUST CLOSE THE PATH AND THEN REOPEN BOTH OF THEM. THERE WILL BE NO FURTHER PROBLEMS WITH THE PATH, ALTERNATE INDEX OR BASE CLUSTER AND THEY WILL FUNCTION AS EXPECTED TO - UNTIL THE ONLINE COMES DOWN AND UP. (AT THAT TIME YOU MUST GO THROUGH THE ODD OPENING ROUTINE I DESCRIBED ABOVE).

I TRIED ALL COMBINATIONS IN THE FCT OF OPEN INITIAL AND OPEN DEFERRED, BUT EACH PAIR PRODUCED THE SAME RESULTS - ONE FILE HAD TO BE CLOSED BEFORE THEY COULD BOTH BE OPENED.

IF YOU HEAR OF A BETTER .SOLUTION (SURELY THERE IS ONE) PLEASE LET ME KNOW AND OF COURSE IF I FIND A BETTER WAY, I'LL CONTACT YOU. BY THE WAY, WE HAVE AN IBM 4341 DOS/VSE.

IF SYS-ED HAS A CLASS BROCHURE, PLEASE SEND IT TO ME AND I'LL BE SURE MY BOSS SEES IT. SHE WAS IMPRESSED WITH YOUR RESPONSE TO THIS MATTER.

TAKE CARE AND THANKS AGAIN,

(Signature on file)

JULIE ARDITTI