One thing I like to do in these types of situations is redo the code.

If this is a home grown program, I would reassemble/lked (compile/lked) on the 
z/OS V2.1 system and try it that way.  Or do a compare between the two versions.

There might be some changes in z/OS V2.1 that are not obvious.

Lizette


> -----Original Message-----
> From: IBM Mainframe Discussion List [mailto:[email protected]]
> On Behalf Of Donald Likens
> Sent: Saturday, June 20, 2015 6:32 AM
> To: [email protected]
> Subject: Wait forever
> 
> We have a situation we do not understand. The code I am sharing was
> developed on a z/OS 1.3 system and has been executed millions of times but
> from what I can tell it simply did not work once on a z/OS 2.1 system. This
> code has not been changed for over a year. This execution ran for a day
> before it went into a wait state and never came out of the wait state (waited
> a day before taking any action). In the dump I see:
> 
> 1 PRB with an interrupt code of 0001 (wait).
> 1 SVRB with an interrupt code 0f 0033 (dump) and then one with an interrupt
> code of 0009 (delete in estae) No IRBs (I think this means no active timers)
> The OPSW in the PRB points to the next instruction following the following
> code:
> 
>                                     523 *C     WAIT FOR WKWAITI SECONDS
>                                     524          STIMER 
> REAL,STIMERX,DINTVL=WKWAITI
> 526+*        MACDATE = 08/19/88
> 000524 4110 B55B            0055B   527+         LA    1,WKWAITI
> 000528 41F0 CC7C            00C7C   528+         LA    15,STIMERX
> 00052C 4100 00B3            000B3   529+         LA    0,179(0,0)
> 000530 8900 0018            00018   530+         SLL   0,24(0)
> 000534 0A2F                         531+         SVC   47
> 000536 D703 CC90 CC90 00C90 00C90   532          XC   STIMECB,STIMECB
>                                     533          WAIT 1,ECBLIST=WAITLIS
>                                     534+*        MACDATE  10/20/88
> 00053C 4100 0001            00001   536+         LA    0,1(0,0)
>                                     538+*
>                                     539+*
> 000540 4110 CC88            00C88   541+         LA    1,WAITLIST
> 000544 1311                         542+         LCR   1,1
> 000546 0A01                         543+         SVC   1
> ===> PSW points here. <===
> 
> WAITLIST= C+C88 = 7C88 = 00007C80 4110F014 0A0207FE 153A9FE4 80007C90
> 000C88                             1270 WAITLIST DS    0F
> 000C88                             1271 ASMNSTOP DS    A
> 000C8C 80                          1272          DC    X'80'
> 000C8D 000C90                      1273          DC    AL3(STIMECB)
> 000C90                             1274 STIMECB  DS    F
> STIMECB = 308FF040       I am assuming un-waited by ABTERM
> 153A9FE4 = 308FF040      I am assuming un-waited by ABTERM
> 
> The STIMER exit follows:
> 000C7C                             1254 STIMERX  DS    0H
>                                    1255 * TIMER EXPIRED
>                                    1256          DROP  R12
>                  R:F  00C7C        1257          USING *,R15
>                                    1258 *C   POST STIMECB
>                                    1259          POST  STIMECB,X'40'
>                                    1260+*        MACDATE  03/11/11
> 000C7C 4100 0040            00040  1262+         LA    0,X'40'(0,0)
> 000C80 4110 F014            00C90  1263+         LA    1,STIMECB
> 000C84 0A02                        1264+         SVC   2
>                                    1265          DROP  R15
>                                    1266 *C   RETURN
> 000C86 07FE                        1267          BR    R14
> 
> I did not see the need to save and restore the registers for this code.
> 
> Does anyone have any ideas or recommendations to help fix this problem?
> 
> Thank you for even looking at this.
> 

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to