TOCPREVNEXTINDEX

Lund Performance Solutions


SOSLOGX Report Card

The Report Card Screen

The Report Card is a standard SOS/3000 report that shows logged SOS/3000 performance data that is summarized in a concise fashion that is suitable for management. There are 18 different options that can be mixed and matched as desired to create a customized report best suited to your needs.
The initial set-up of the Report Card is very simple. It requires you to select options in the Report Card configuration submenu. That submenu is accessible in two ways:
  • While you are in SOS/3000:
  • Press SCREEN MENU (F7) function key;
  • Press L for LOG TRENDS PROGRAM;
  • Press OPTIONS (F1) function key.
  • Type 9 to chose Report card configuration (SUBMENU). Press Enter.
  • From an MPE prompt:
  • Run SOSLOGX.PUB.LPS and press the OPTIONS (F1) function key.
  • Type 9 to chose Report card configuration (SUBMENU). Press Enter.
  • Then you will see a Report Card Configuration Menu as is shown in Figure 28.1.
    Each of the Report Card Configuration Options is designed for ease of use and flexibility so that you may tailor the reports specifically for your system.


    Figure 28.1 SOSLOGX Report Card configuration submenu

    Report Card Configuration Options

    Lines per output page (option 1)
    This option allows you to configure the output of your reports to your line printer. Enter the number of lines here that your printer prints per page. The default value is 60 lines per page.
    Global CPU Statistics section (option 2)
    Your choices are:
  • 0 - Don’t print
  • 1 - Print on same page
  • 2 - Print on new page
  • If you choose either 1 or 2, this report will print and it will average the information for the time frame that you choose when you are in REPORT CARD. To view this report, see "Viewing the Report Card". The mean, standard deviation, high value and low value for each of the statistics will be listed (see Figure 28.2).

    07/16/01
    SOS/3000 Report Card
    Page 1
     
     
     
     
     
    Item
    Mean
    SD
    High
    Low
     
     
     
     
     
    *** Global CPU Statistics ***
    Total busy
    86.2
    8.07
    98.9
    68.7
    AQ CPU%
    1.5
    1.37
    2.9
    .0
    BQ CPU%
    9.5
    3.06
    16.3
    4.4
    CQ CPU%
    26.0
    5.68
    38.2
    19.4
    DQ CPU%
    35.9
    12.85
    58.8
    18.2
    EQ CPU%
    .0
    .00
    .0
    .0
     
     
     
     
     
    Memory Manager%
    5.1
    1.70
    7.9
    2.7
    Dispatcher%
    .1
    .04
    .2
    .1
    ISC/OH%
    .9
    1.83
    4.6
    .0
    Pause for Disc I/O%
    12.9
    7.97
    31.3
    1.1
    Idle%
    .9
    1.83
    4.6
    .0
     
     
     
     
     
    CPU Queue Length
    3
    .5
    4
    2
    Launches/sec
    70
    11.4
    88
    52
    CPU CM%
    26
    7.5
    40
    16
    SAQ
    200
    .0
    200
    200

    Figure 28.2 Report Card: Sample Global CPU Statistics Report
    Global Misc Statistics section (option 3)
    Your choices are:
  • 0 - Don’t print
  • 1 - Print on same page
  • 2 - Print on new page
  • If you choose either 1 or 2, this report will print and it will average the information for the time frame that you choose when you are in REPORT CARD. When you choose "2 - Print on new page", then this report option will print on a different page than the Global CPU Statistics Report. To view this report, see "Viewing the Report Card". The mean, standard deviation, high value and low value for each of the statistics will be listed (see Figure 28.3).

    07/16/01
    SOS/3000 Report Card
    Page 2
     
     
     
     
     
     
    Item
    Mean
    SD
    Conf.
    High
    Low
     
     
     
     
     
     
    *** Global Miscellaneous Statistics ***
    Session
    102
    1.9
    5
    105
    99
    Jobs
    15
    .6
    5
    16
    14
    Processes
    629
    9.8
    5
    655
    621
    CM to NM Switches/sec
    483
    433.8
    5
    1553
    225
    NM to CM Switches/sec
    30
    11.3
    5
    56
    19
    Transactions/minute
    498
    249.8
    5
    1012
    320
    First Response
    .1
    .03
    5
    .2
    .1
    Prompt Response
    .3
    .08
    5
    .4
    .2

    Figure 28.3 Report Card: Sample Global Miscellaneous Statistics Report
    Global Memory Statistics section (option 4)
    Your choices are:
  • 0 - Don’t print
  • 1 - Print on same page
  • 2 - Print on new page
  • If you choose either 1 or 2, this report will print and it will average the information for the time frame that you choose when you are in REPORT CARD. When you choose "2 - Print on new page", then this report option will print on a new page. To view this report, see "Viewing the Report Card". The mean, standard deviation, high value and low value for each of the statistics will be listed (see Figure 28.4.)

    07/16/01
    SOS/3000 Report Card
    Page 3
     
     
     
     
     
     
    Item
    Mean
    SD
    Conf.
    High
    Low
     
     
     
     
     
     
    *** Global Memory Statistics ***
    Page Faults/sec
    14
    4.4
    5
    23
    6
    Library Fault %
    2
    1.6
    5
    5
    1
    Memory Cycles
    1
    .6
    5
    2
    0
    Read Hit %
    94
    3.6
    5
    99
    87
    Overlay Rate
    6
    5.8
    5
    18
    1
    Swaps/Launch
    .35
    .074
    5
    .51
    .27

    Figure 28.4 Report Card: Sample Global Memory Statistics Report
    Disc Statistics section (option 5)
    Your choices are:
  • 0 - Don’t print
  • 1 - Print on same page
  • 2 - Print on new page
  • If you choose either 1 or 2, this report will print and it will average the information for the time frame that you choose when you are in REPORT CARD. When you choose "2 - Print on new page", then this report option will print on a new page. To view this report, see "Viewing the Report Card". The mean, standard deviation, high value and low value for each of the statistics will be listed (see Figure 28.5.)
    This option works with menu option:
    12) Break down disc stats by ldev (Y)
    The default for this option is Y (Yes). If you choose to change this to N (No), only a Total Global Disc Statistics Report will print.
    Notice that each device will be followed by a report on that device and then at the end a Total Global Disc Statistics summing all of the disc drives will print. For this sample we have labeled the device as "n" (some integer).

    07/16/01
    SOS/3000 Report Card
    Page 4
     
     
     
     
     
     
    Item
    Mean
    SD
    Conf.
    High
    Low
     
     
     
     
     
     
    *** Global Disc Statistics ***
    Disc Reads/Sec
    3
    .8
    5
    4
    2
    Disc Writes/Sec
    3
    .8
    0
    4
    2
    Disc IO/Sec
    6
    1.3
    5
    8
    4
    Disc Request Queue Length
    1.92
    .901
    5
    3.36
    .61
    Percent of Total Disc I/O
    14
    3.4
    5
    22
    11
    Total I/O
    1701
    388.5
    5
    2472
    1233

    Figure 28.5 Report Card: Sample Global (All) Disc Statistics Report
    Stop Reasons section (option 6)
    Your choices are:
  • 0 - Don’t print
  • 1 - Print on same page
  • 2 - Print on new page
  • If you choose either 1 or 2, this report will print and it will average the information for the time frame that you choose when you are in REPORT CARD. When you choose "2 - Print on new page", then this report option will print on a different page. To view this report, see "Viewing the Report Card". The mean, standard deviation, high value and low value for each of the statistics will be listed (see Figure 28.6).
    Remember, Stop Reasons are the reasons that a process loses the CPU as determined by the MPE dispatcher. This will report the Stop Reasons encountered during the interval on which you are reporting. It is possible to list all Stop Reasons encountered or not by choosing option:
    14) Suppress unencountered stop reasons (N)

    07/16/01
    SOS/3000 Report Card
    Page 5
     
     
     
     
     
     
    Item
    Mean
    SD
    Conf.
    High
    Low
     
     
     
     
     
     
    *** Stop Reasons ***
    Terminal Read Stop %
    12
    3.9
    5
    19
    8
    Terminal Write Stop %
    0
    .3
    5
    1
    0
    Other I/I Stop %
    1
    .7
    5
    3
    1
    IOWAIT Stop %
    1
    .0
    5
    1
    1
    Suspend on Father Stop %
    0
    .3
    5
    1
    0
    Suspend on Son Stop %
    1
    .3
    5
    2
    1
    Port Stop %
    11
    4.2
    5
    19
    7
    Junk Stop %
    1
    .3
    5
    2
    1
    Fill Disk Stop %
    1
    .7
    5
    2
    0
    CPU Preempt Stop %
    13
    2.6
    5
    18
    9

    Figure 28.6 Report Card: Sample Stop Reasons Report
    Workload Summary section (option 7)
    Your choices are:
  • 0 - Don’t print
  • 1 - Print on same page
  • 2 - Print on new page
  • If you choose either 1 or 2, this report will print and it will average the information for the time frame that you choose when you are in REPORT CARD. When you choose "2 - Print on new page", then this report option will print on a different page. To view this report, see "Viewing the Report Card". The mean, standard deviation, high value and low value for each of the statistics will be listed (see Figure 28.7).
    A report will print for each workload that you have defined in your workload definition file, SOSWKDEF (see "Defining Workloads"). If no workloads have been defined then only three workloads will be reported. These three are the default workloads of JOBS, SESSIONS, and SYSPROCS (system processes).
    If you have changed your workload definition file (SOSWKDEF) during the period you are including within your report, Report Card will break your report into time slices for every change to the workload definition file. In each of these sections an informational message will be printed noting that multiple workloads have been detected. This ensures that the data is reliable because averaging data that was not created with the same criteria is like mixing apples and oranges.
    For this sample we have displayed the "Sessions" workload summary.

    07/16/01
    SOS/3000 Report Card
    Page 6
     
     
     
     
     
     
    Item
    Mean
    SD
    Conf.
    High
    Low
     
     
     
     
     
     
    *** Sessions Workload Summary ***
    CPU Percentage
    19.3
    4.82
    5
    31.4
    14.5
    Total Disk I/O %
    40.8
    6.02
    5
    49.5
    31.0
    Average Prompt Response Time
    .3
    .05
    5
    .3
    .2
    Total Transactions
    1328
    180.7
    5
    1665
    1085
    CPU Time per Transaction
    46
    16.6
    5
    89
    30
    Disc I/O per Transaction
    4
    1.0
    5
    5
    2
    Average First Response Time
    .1
    .03
    5
    .2
    .1
    CPU Time in msecs
    588830
    148370.4
    5
    965530
    449680
    CPU CM Time in msecs
    136880
    42720.3
    5
    196350
    74080
    Disc Read Count
    3931
    844.8
    5
    5709
    2883
    Disc Write Count
    843
    295.1
    5
    1263
    455

    Figure 28.7 Report Card: Sample Workload Summary Report
    Average Advice section (option 8)
    Your choices are:
  • 0 - Don’t print
  • 1 - Print on same page
  • 2 - Print on new page
  • If you choose either 1 or 2, this report will print and it will average the information for the time frame that you choose when you are in REPORT CARD. When you choose "2 - Print on new page", then this report option will print on a different page. These are the informational or excessive messages that occurred during the average of all performance indicators, in other words, it is a representative time performance report of your system (see Figure 28.8).
    The Mean Interval Advice messages are a useful indicator of what the average performance hindrances might be for your system.
    There is also an option not to report certain messages. For example, if you are not interested in reporting the MODERATE messages choose option:
    15) First advice type to suppress (3) (See this menu option for more details).
    Your choices are:
  • 1 - EXCESSIVE advice messages
  • 2 - HEAVY advice messages
  • 3 - MODERATE advice messages

  • 07/16/01
    SOS/3000 Report Card
    Page 7
     
     
     
     
     
     
    *** Mean Interval Advice ***
    Comp. Mode to Native Mode Switch rate during this interval was EXCESSIVE
    <GE03>
    Disc I/O indicator #1 (CPU Pause Disc) reveals a HEAVY I/O Bottleneck
    <DE01>
    Memory indicator #2 (CPU Memory) reveals a MODERATE memory load
    <ME02>
    Memory indicator #4 (Page Fault Rate) reveals a MODERATE memory load
    <ME04>

    Figure 28.8 Report Card: Sample Mean Interval Advice Report
    Peak Advice section (option 9)
    Your choices are:
  • 0 - Don’t print
  • 1 - Print on same page
  • 2 - Print on new page
  • If you choose either 1 or 2, this report will print and it will select the highest or worst measurement information for the time frame that you choose when you are in REPORT CARD. When you choose "2 - Print on new page", then this report option will print on a different page. Peak Advice Messages are the informational or excessive messages that occured during the highest or worst measurement of all the performance indicators. In other words, it is a snapshot of when your HP 3000 resources are being stressed the most (see Figure 28.9).
    The Peak Interval Advice messages are a useful indicator of what happens to your HP 3000 during high resource utilization periods.
    There is also an option not to report certain messages. For example if you are not interested in reporting the MODERATE messages choose option:
    15) First advice type to suppress (3) (See this menu option for more details).
    Your choices are:
  • 1 - EXCESSIVE advice messages
  • 2 - HEAVY advice messages
  • 3 - MODERATE advice messages

  • 07/16/01
    SOS/3000 Report Card
    Page 7
     
     
     
     
     
     
    *** Peak Interval Advice ***
    CPU consumption due to system overhead during this interval was MODERATE
    <GE01>
    Native Mode to Comp. Mode Switch rate during this interval was MODERATE
    <GE02>
    Disc I/O indicator #1 (CPU Pause Disc) reveals an EXCESSIVE I/O Bottleneck
    <DE01>
    Memory indicator #2 (CPU Memory) reveals a MODERATE memory load
    <ME02>
    Memory indicator #3 (Swap/Launch) reveals a MODERATE memory load
    <ME03>
    Memory indicator #4 (Page Fault Rate) reveals a MODERATE memory load
    <ME04>

    Figure 28.9 Report Card: Sample Peak Interval Advice Report
    Advice Detail Report section (option 10)
    Your choices are:
  • 0 - Don’t print
  • 1 - Print on same page
  • 2 - Print on new page
  • This option works with submenu options 15-18.
    If you choose either 1 or 2, this report will print all informational messages that occurred during the time frame that you choose when you are in REPORT CARD. When you choose "2 - Print on new page", then this report option will print on a different page. A sample report will appear displaying all Advice Detail Messages (see Figure 28.10).
    We advise you that you choose a very short time interval for your REPORT CARD if you choose to print this option, otherwise you will probably be printing reams of paper.
    This option is very useful if you have a specific time interval in mind that you want to investigate. For example, if you notice that a batch job was very slow at a certain time you can list the informational messages to investigate the cause.
    There is also an option not to report certain messages. For example if you are not interested in reporting the MODERATE messages choose option:
    15) First advice type to suppress (3) (See this menu option for more details).
    Your choices are:
  • 1 - EXCESSIVE advice messages
  • 2 - HEAVY advice messages
  • 3 - MODERATE advice messages
  • You also have the option of ordering your advice detail by message if you choose submenu option:
    18) Report informational advice messages (N) or (Y) (See this menu option for more details).
    Another option is to report informational advice messages not just exceptional messages. Choose submenu option:
    18) Report informational advice messages (N) or (Y) (See this menu option for more details).

    07/16/01
    SOS/3000 Report Card
     
     
     
     
     
    *** Advice Detail Report ***
     
     
     
     
     
    Disc I/O indicator #1 (CPU Pause Disc) reveals a MODERATE I/O Bottleneck
    <DE01>
     
    07/16/01 06:30 AM
    07/16/01 06:35 AM
    07/16/01 06:45 AM
    07/16/01 06:55 AM
     
    07/16/01 07:00 AM
    07/16/01 07:20 AM
    07/16/01 08:20 AM
    07/16/01 08:50 AM
     
    07/16/01 08:55 AM
    07/16/01 09:20 AM
    07/16/01 10:30 AM
    07/16/01 11:25 AM
     
    07/16/01 11:55 AM
    07/16/01 01:20 PM
    07/16/01 02:40 PM
    07/16/01 02:55 PM
     
    07/16/01 03:15 PM
    07/16/01 04:50 PM
    07/16/01 04:55 PM
    07/16/01 05:25 PM
    Disc I/O indicator #1 (CPU Pause Disc) reveals a HEAVY I/O Bottleneck
    <DE01>
     
    07/16/01 06:20 AM
    07/16/01 06:50 AM
    07/16/01 08:10 AM
    07/16/01 08:30 AM
     
    07/16/01 10:10 AM
    07/16/01 11:30 AM
    07/16/01 02:20 PM
    07/16/01 02:45 PM
     
    07/16/01 03:35 PM
    07/16/01 04:20 PM
    07/16/01 04:45 PM
     
    Disc I/O indicator #1 (CPU Pause Disc) reveals an EXCESSIVE I/O Bottleneck
    <DE01>
     
    07/16/01 07:45 AM
    07/16/01 09:15 AM
    07/16/01 09:20 AM
    07/16/01 09:40 AM
     
    07/16/01 10:20 AM
    07/16/01 10:25 AM
    07/16/01 01:25 PM
    07/16/01 01:40 PM
     
    07/16/01 02:50 PM
    07/16/01 03:05 PM
    07/16/01 04:00 PM
    07/16/01 04:10 PM
     
    07/16/01 04:25 PM
    07/16/01 05:00 PM
    07/16/01 05:05 PM
    07/16/01 05:10 PM
     
    07/16/01 05:15 PM
     
     
     

    Figure 28.10 Report Card: Sample Advice Detail Report
    Advice Count section (option 11)
    Your choices are:
  • 0 - Don’t print
  • 1 - Print on same page
  • 2 - Print on new page
  • If you choose either 1 or 2, this report will print a count of all informational messages that occurred during the time frame that you choose when you are in REPORT CARD. When you choose "2 - Print on new page", then this report option will print on a different page. A sample report will appear displaying the count on all advice messages (see Figure 28.11).
    This option is very useful to indicate which informational messages occur most often. An example of this is if you see the message "Disc I/O indicator #1 (CPU Pause Disc) reveals an EXCESSIVE I/O Bottleneck" with a count of 300 and you only have 300 samples, you should investigate this to determine if you truly have a resource problem or an application problem.
    There is also an option not to report certain messages. For example, if you are not interested in reporting the MODERATE messages choose option:
    15) First advice type to suppress (3) (See this menu option for more details).
    Your choices are:
  • 1 - EXCESSIVE advice messages
  • 2 - HEAVY advice messages
  • 3 - MODERATE advice messages

  • 07/16/01
    SOS/3000 Report Card
    Page 7
     
     
     
     
     
     
    *** Advice Counts Report ***
    Disc I/O indicator #1 (CPU Pause Disc) reveals a MODERATE I/O Bottleneck
    <DE01>
    Count:
    22
     
     
     
     
    Disc I/O indicator #1 (CPU Pause Disc) reveals a HEAVY I/O Bottleneck
    <DE01>
    Count:
    18
     
     
     
     
    Disc I/O indicator #1 (CPU Pause Disc) reveals an EXCESSIVE I/O Bottleneck
    <DE01>
    Count:
    22
     
     
     
     
    Native Mode to Comp. Mode Switch rate during this interval was MODERATE
    <GE02>
    Count:
    73
     
     
     
     
    Native Mode to Comp. Mode Switch rate during this interval was HEAVY
    <GE02>
    Count:
    19
     
     
     
     
    Native Mode to Comp. Mode Switch rate during this interval was EXCESSIVE
    <GE02>
    Count:
    5
     
     
     
     
    Comp. Mode to Native Mode Switch rate during this interval was MODERATE
    <GE03>
    Count:
    10
     
     
     
     
    Comp. Mode to Native Mode Switch rate during this interval was HEAVY
    <GE03>
    Count:
    10
     
     
     
     
    Comp. Mode to Native Mode Switch rate during this interval was EXCESSIVE
    <GE03>
    Count:
    115
     
     
     
     
    Global average response time during this interval was MODERATE
    <GE04>
    Count:
    3
     
     
     
     
    Global average response time during this interval was HEAVY
    <GE04>
    Count:
    1
     
     
     
     
    Global average response time during this interval was EXCESSIVE
    <GE04>
    Count:
    1
     
     
     
     
    Memory indicator #3 (Swap/Launch) reveals a MODERATE memory load
    <ME03>
    Count:
    24
     
     
     
     
    Memory indicator #3 (Swap/Launch) reveals a HEAVY memory load
    <ME03>
    Count:
    7
     
     
     
     
    Memory indicator #4 (Page Fault Rate) reveals a MODERATE memory load
    <ME04>
    Count:
    24
     
     
     
     
    Memory indicator #4 (Page Fault Rate) reveals a HEAVY memory load
    <ME04>
    Count:
    50
     
     
     
     
    Memory indicator #4 (Page Fault Rate) reveals an EXCESSIVE memory load
    <ME04>
    Count:
    28
     
     
     
     

    Figure 28.11 Report Card: Sample Advice Count Report
    Break down disc stats by ldev (option 12)
    This option (option 12) works with menu option:
    5) Disc Statistics section
    If option 5 is set to print and option 12 is set to the default of Y (Yes) is used, then each and every LDEV will be followed by a report on that device and then at the end a Total Global Disc Statistics summing all of the disc drives will print.
    If you choose to change this to N (No) only a Total Global Disc Statistics Report will print, summing the average disc statistics.
    Report by zone (option 13)
    Zones or shifts are originally defined in the Data Break Configuration submenu (see "Data break configuration menu (SUBMENU)"). The default is to accept the zones as you have previously defined them and then each zone is reported individually but summed in the global statistics reports.
    If you choose to ignore the zones that you have defined (change this option to N (No)), then the entire time period you specify will be examined as a whole and averaged together.
    This option is very useful if you have specified zones that indicate different usage/utilization of your HP 3000. An example of this is to "zone" the heavy interactive use from the heavy batch processing use. Are the interactive users heavy from 9 am to 3 pm? Is this different from the batch job runs?
    Suppress unencountered stop reasons (option 14)
    The default value for this option (14) is Y (Yes). This option works with menu option:
    6) Global Stop Reasons section
    Remember, Stop Reasons are reasons that a process loses the CPU as determined by the MPE dispatcher. This will report the Stop Reasons encountered during the interval that you are reporting on.
    If you would like a listing of all the possible Stop Reasons, whether or not they were encountered, change the value for this option to N (No).
    First advice type to suppress (option 15)
    This menu option works with all Advice menu options (options 8-11).
    Your choices are:
  • 1 - EXCESSIVE advice messages
  • 2 - HEAVY advice messages
  • 3 - MODERATE advice messages
  • The default value for this option is blank. If you would like to exclude certain types of advice messages choose the associated number. If you choose a number then option "16) Second advice type to suppress" will become available for you to select.
    Second advice type to suppress (option 16)
    If you choose to exclude a type of Advice message then it will not appear on any of your reports. An example of this is if you choose to exclude the MODERATE and HEAVY messages from your reports because you were only interested in EXCESSIVE message. First you would choose menu option 15 and enter a 2 to exclude HEAVY advice messages. The resulting reports that you choose would not have HEAVY or MODERATE advice messages on your Advice Reports.
    Notice that this menu option will not be available if you do not choose to print any Advice Reports.
    This menu option is available if you have chosen to print at least one of the Advice Reports and you entered a valid value (1, 2 or 3) in option 15 - First advice type to suppress.
    Report advice detail by message (option 17)
    This option works in conjunction with submenu option 10 - Advice Detail Report section.
    If you choose to print the Advice Detail then you have the option of ordering your detail report by advice message. you will see a detail report that uses the advice message as a header and then notes the time stamp when that message occurred. This is an easy way to scan your data for peak resource stress time periods. Perhaps segregating them as zones will work very well for your shop.
    Report informational advice messages (option 18)
    This option works in conjunction with submenu option 10 - Advice Detail Report section
    If you choose to print the Advice Detail, then you have the option of suppressing or enabling the reporting of information messages as well as exceptional messages.

    How to Run Report Card

    The way to access Report Card is to run SOSLOGX.PUB.LPS and press UTILITY KEYS (F5) and the REPORT CARD (F6) function key. You will then see a Report Card time prompt screen. Choose the dates and times you would like your reports to include.
    Pressing the REPORT CARD function key will generate a Report Card. To view the report card, see "Viewing the Report Card."

    Viewing the Report Card

    To print the Report Card to your screen:
  • Exit SOSLOGX by pressing EXIT SOSLOGX (F8).
  • At the MPE prompt, type listspf.
  • Scan the list of spoolfiles. The last RCRDLIST file in the list is the most recent printout of the Report Card.
  • At the MPE prompt, type printo, then the number associated with the file, minus the number sign (#) and the zero (0). In Figure 28.12, the file is number 24496.


  • Figure 28.12 Sample listspf screen and printo command

    Lund Performance Solutions
    www.lund.com
    Voice: (541) 812-7600
    Fax: (541) 812-7611
    info@lund.com
    TOCPREVNEXTINDEX