BMC24529 Outgrp <nnn> has lost <nnnn> trace records

Version 2
    Share This:

    This document contains official content from the BMC Software Knowledge Base. It is automatically updated when the knowledge article is modified.


    PRODUCT:

    MainView for DB2


    APPLIES TO:

    Apptune for DB2, Mainview for DB2 Management, BMC Performance for SQL, System and SQL Performance for DB2 (all supported versions)



    QUESTION:

    How to resolve BMC24529 Outgrp <nnn> has lost <nnnn> trace records messages?
    BMC24529 Outgrp <nnn> has lost <nnnn> trace records message is logged when the Data space buffers become completely full. While the NGL is in the retry process, the continued production of IFCIDs will cause the Data space buffers to be filled up. The "lost" messages means that the high volume IFCID production outran NGL and sustained long enough that all the buffers were filled up and data was permanently lost.


     


    ANSWER:

     

    Solution: Increase "Max log buffers" and "Dataspace buffer size" in the DOMPLEX Output Group definition.
    Note: The sum of data space for all output groups assigned to the same data collector cannot exceed 2000MB.  The total dataspace size in use is displayed in the DBC JESMGLG: BMC24735 Total DSSIZE used for all Output Groups is  235Mb.

    Instructions to increase Max log buffers and Dataspace buffer size in DOMPLEX option set:
    1) Start a product session to navigate to Administration > DOMPLEX Option Sets. For detailed instructions. click Accessing the Online interface.
    2) Modify the Output Group to change Max log buffers and Dataspace buffer size . For detailed instructions, click Modifying the Output Groups.

      

    + DOMPLEX Parameters           Parameters that apply to entire DOMPLEX  
    + Data Collector List (1)          Data Collector(DBC) subsystems in DOMPLEX  
    + DB2 Monitor List (11)           DB2 Sub-systems to be monitored  
    - OutGp DCID DspSize (6)        Output Groups - valid range: 001-256 
      - 011 N11J 20                         Dataspace buffer size: 1-2000 (MB)  
        + Data Classes                        Specify IFCIDs to be stored in this group
        - NGL LOGSET Parameters       LOGSET attributes used by this group     
          Logset time span . . . . . . . .  7D       (nD,nH,nM)                   
          Max log buffers. . . . . . . . .  10       (2-20)                       
          Max read buffers . . . . . . . .  4        (2-99)                       
          Deferred write time. . . . . . .  60       (1-999 sec.)                 
          Minimum log file data sets (LDS)  2        (1-99)                       
          Maximum log file data sets (LDS)  5        (1-99)                       
          Space to allocate (per LDS). . .  100      (1-9999 MB) note: 1 CYL=720KB

      

    3) Click for Instructions for the NGL Output group changes to take effect.
    4) Verify the changes made Output Group.

      
       
    • See DBC JESMSGLG message BMC24524 DC01 Outgrp 001 trk  nnn of  yyy in DATASPACE to verify Dataspace changes.

    •  
    • Enter console command /dcssid NGL,CMD,piid,DISPLAY LOGSET logset-name to verify changes to NGL LOGSET Parameters. e.g /DC01 NGL,CMD,DOM1,DISPLAY LOGSET DOM1L001 for output group 001.

      

    5) If high activity persists long enough then data will be lost no matter how much buffer space is dedicated. If this is the case, then create more Output Groups and separate the data so that there are more write engines working in parallel. Refer to Creating an Output group.

       Related Knowledge Articles:
    BMC24563 Logset logging delayed, all buffers full
    How to increase the thread history available in MVDB2 Data Collector active trace data sets?

     


    Article Number:

    000139192


    Article Type:

    FAQ/Procedural



      Looking for additional information?    Search BMC Support  or  Browse Knowledge Articles