IBM Z and LinuxONE IBM Z

  • 1.  Does anyone know how to get out of an out-of-spool condition?

    Posted Mon October 18, 2021 09:49 AM
    I am not a great sysprog.  I run my own system on IBM's Dallas Vendor systems, but not very well.  I had a job that did excessive spool use through WTO's that filled the system log.   Now I can't log on to do anything, and I don't know how to purge/recycle.  I know that I can send JES2 commands through the console that I have in the VM that is hosting my guest.   I just don't know JES commands that well and am slogging thru the docs.   I imagine that I am not the first person to have painted myself into this particular corner.

    Thanks,

     $HASP003           OUTPUT (AS DEFINED BY CONDEF DISPMAX)                   
 $HASP355 SPOOL VOLUMES ARE FULL                                            
 H *$HASP050 JES2 RESOURCE SHORTAGE OF TGS - 100% UTILIZATION REACHED       
 $HASP355 SPOOL VOLUMES ARE FULL                                            
 H *$HASP050 JES2 RESOURCE SHORTAGE OF TGS - 100% UTILIZATION REACHED       
 $HASP355 SPOOL VOLUMES ARE FULL                                            
 $HASP355 SPOOL VOLUMES ARE FULL                                            
 H *$HASP050 JES2 RESOURCE SHORTAGE OF TGS - 100% UTILIZATION REACHED       
 $HASP355 SPOOL VOLUMES ARE FULL                                            
 H *$HASP050 JES2 RESOURCE SHORTAGE OF TGS - 100% UTILIZATION REACHED       
 $HASP355 SPOOL VOLUMES ARE FULL                                            
 H *$HASP050 JES2 RESOURCE SHORTAGE OF TGS - 100% UTILIZATION REACHED       
 $HASP355 SPOOL VOLUMES ARE FULL                                            
 H *$HASP050 JES2 RESOURCE SHORTAGE OF TGS - 100% UTILIZATION REACHED       
 $HASP355 SPOOL VOLUMES ARE FULL                                            
 H *$HASP050 JES2 RESOURCE SHORTAGE OF TGS - 100% UTILIZATION REACHED       
 $HASP355 SPOOL VOLUMES ARE FULL                                            
 H *$HASP050 JES2 RESOURCE SHORTAGE OF TGS - 100% UTILIZATION REACHED       ​


    ------------------------------
    Joseph Devlin
    ------------------------------


  • 2.  RE: Does anyone know how to get out of an out-of-spool condition?

    IBM Champion
    Posted Mon October 18, 2021 11:20 AM
    Hi Joe,

    The first thing I'd try is $POJQ,AGE>n where n is days. Try 7 to purge anything larger than a week.

    Second thing: I don't know if Dallas activates the JES2 emergency subsystem, but starting with 2.mumble there's a new emergency subsystem for this situation. You need to log on to TSO to that subsystem; I don't know if Dallas has USS set up to allow logging on to an alternate subsystem (if not, they should); I haven't checked our ADCD image lately.

    Cheers,
    Ray

    ------------------------------
    Ray Mullins
    Principal Software Engineer
    Rocket Software, Inc.
    Roseville CA
    ------------------------------

    Original Message


  • 3.  RE: Does anyone know how to get out of an out-of-spool condition?

    Posted Tue October 19, 2021 08:11 AM
    Make sure you stop/cancel whatever is issuing the WTO's before purging items >7 days.... or you will likely be back in the same boat soon.


    Robert Barringer
    Systems Architect - NCMMIS / eMedNY

    www.gdit.com 

    GDIT





    Original Message


  • 4.  RE: Does anyone know how to get out of an out-of-spool condition?

    Posted Tue October 19, 2021 12:13 PM
    Hi, 

    Thanks for all your help.  I will try to not make so many WTO's in the future.    I probably foolishly assumed that SYSLOG's use of spool files would act as a circular buffer if things got full.    I am sure that there are important reasons, that only true mainframers know, why it is more robust to have the system seize up entirely rather than discard old messages.    Maybe all of the cool kids have automation in place that effectively does this.  I manage my own guest at IBM Dallas and am definitely not one of the cool kids.

    ------------------------------
    Joseph Devlin
    ------------------------------

    Original Message


  • 5.  RE: Does anyone know how to get out of an out-of-spool condition?

    IBM Champion
    Posted Tue October 19, 2021 08:40 AM
    Hi Joe,

    First, apply the command to display the status of spool volumes: $DSPL
    To clean up all spools older than 7 days issue command: $POJOBQ,ALL,PROTECTED,DAYS>7
    Automatic cleanup of syslog output can be done to help prevent the spool space full condition. This can be done by placing the following command at the bottom of the JES2 parameter deck: $TA,T=00.30,I=86400,'$POJOBQ,Q=C,DAYS>7' (This will delete any class C output every 24 hours at 00:30 AM)
    Be sure that Class C output is the normal class specified in IEASYSLV member of the PARMLIB for the syslog output class. The command will need to be modified as well if another class is used.

    Maybe you can use the following command $D JOBQ,SPOOL=(%>5) before to determine if some job uses more than 5% of the spool to manually delete these jobs.

    Regards

    ------------------------------
    Guillaume VENTRE
    ------------------------------

    Original Message


  • 6.  RE: Does anyone know how to get out of an out-of-spool condition?

    Posted Wed January 03, 2024 02:28 PM

    Hi Guillaume,

    I'm interested in your $TA timer statement but had a couple of questions:

    1) Would "$T A,ALL" display the command after you set it?

    2) Once you've got your timer command set, how would you delete it?

    Thanks for any help, all the info here was very useful.

    Keith



    ------------------------------
    keith shaffer
    ------------------------------

    Original Message


  • 7.  RE: Does anyone know how to get out of an out-of-spool condition?

    Posted Thu January 04, 2024 10:24 AM
    Also: Check out the JES2 Emergency Subsystem.

    SPOOL FULL happens to us all from time-to-time.  You can end up in a catch-22 situation where you need to logon to TSO, but you can't because the SPOOL is full.  You may need to cancel the running job that is filling up the SPOOL every time you delete some output, but which job?  The JES2 Emergency subsystem can be used to reserve some spool space so you can logon to TSO even in SPOOL FULL.  But you need to practice using it in advance to be sure you have everything set up correctly and know how to use it.  Do a search on "z/os jes2 emergency subsystem".  Marne Walle and Tom Wasik have both done SHARE presentations that talk about this.

    By the way: An IPL in a SPOOL FULL condition is not a good idea.  You may end up needing a COLD START that deletes everything.

    Dave Crow

      



    Original Message


  • 8.  RE: Does anyone know how to get out of an out-of-spool condition?

    Posted Thu January 04, 2024 06:49 AM
    Edited by Paul Newton Fri January 05, 2024 10:38 AM

    All answers are excellent --- I can only add .... https://www.ibm.com/docs/en/zos/2.2.0?topic=section-c-cancel-automatic-commands  << to delete (technically cancel) set JES2 timer actions

    Addendum ... $D ESTLNCT ...... review JES2 estimated line count is enable to cancel jobs that exceed chosen value 

    https://www.ibm.com/docs/en/zos/2.1.0?topic=crs-t-estlnct-control-default-print-line-count-per-job



    ------------------------------
    Paul Newton
    Consulting IT Specialist
    IBM
    Coppell TX
    ------------------------------

    Original Message


  • 9.  RE: Does anyone know how to get out of an out-of-spool condition?

    Posted Tue October 19, 2021 09:10 AM

    Hi Joseph,
    In addition to the command Ray gave you to find out the age of the outputs in spool, you can use "$DJQ,SPL=(%>1)" to find out big output. Im in mainframe operations for some years now, and it is a great tool to identify the source of the shortage, I had times when a single job in a loop has 30% or more of the total spool with it.

    Good luck!



    ------------------------------
    Nicolas Pablo Mulfetti
    ------------------------------

    Original Message