This chapter describes how to use the facilities provided for monitoring CallPath/CICS. It explains:
The administration subsystem monitors the status of your switches, the status of communications with each switch, and the status of the administration subsystem itself.
The request handler and an event handler provide the communication paths to a switch. A single request handler processes the program calls to all attached CallPath SwitchServer/2s, but CallPath/CICS has one event handler for each CallPath SwitchServer/2.
Table 13 indicates with a Yes
which status is possible for each component.
Table 13. Possible Statuses of CallPath/CICS System Components
To display the status of your system:
On the System Administration Menu, select System status information to see the current status of the administration subsystem and all the switches defined to your system.
For each switch you can see the status of the request handler, the event handler, and the switch itself. In Figure 12, each switch is at a different stage in the sequence that occurs when you make a request to start a switch.
The Switch Status column shows the status of communications between CallPath SwitchServer/2 and the switches. This status will not be current if the event handler and request handler status is "inactive".
Figure 12. System Status Information Panel
EQZMD01 System Status Information Subsystem status . . . . . . . . ACTIVE Switch Request Handler Event Handler Switch Name Status Status Status SWITCH01 START REQUESTED INACTIVE INACTIVE SWITCH02 STARTING INACTIVE INACTIVE SWITCH03 ACTIVE START REQUESTED INACTIVE SWITCH04 ACTIVE STARTING INACTIVE SWITCH05 ACTIVE ACTIVE INACTIVE SWITCH06 ACTIVE ACTIVE ACTIVE F1=Help F3=Exit F5=Refresh F7=Backward F8=Forward F11=Panel ID F12=Cancel |
The administration subsystem maintains the following types of log files:
| Type of Log | File Names | |
|---|---|---|
| Message | EQZLOGA | EQZLOGB |
| Trace | EQZTRCA | EQZTRCB |
| Traffic | EQZTRFA | EQZTRFB |
The administration subsystem uses two files for logging each type of information. For each type of log file, only one file is written to at any time. Logging automatically swaps to the alternative file when the administration subsystem starts and if the current file becomes full:
The number of archive files and their file names are defined at installation. To interpret the information contained in an archive file, you need to understand the way that the administration subsystem formats log records. The format of the records written to each type of log is described in Appendix C, "Log Record Formats".
You can swap logging from an active to an alternative file at any time using the Log File Control panel. You can do this to free an active log file for analysis, if necessary.
To swap from an active to an alternative log file:
You see the current file for each log type and its status.
Figure 13. Log File Control Panel
EQZMJ01 Log File Control
Type / next to the log types of the files you want to swap.
Then press Enter.
Log type Current file Status
_ Message EQZLOGA ACTIVE
_ Trace EQZTRCB SWAP REQUESTED
_ Traffic EQZTRFA SWAPPING
F1=Help F3=Exit F5=Refresh F11=Panel ID F12=Cancel
|
The status of a log file can be one of:
Logging is then swapped to the alternative file. You see the name of the new log file in the Current File column. Its status is active.
You can choose to write messages either to message log files or to user-specified transient data queues, using the severity of the data as the selection criterion. You use SCT options to specify which message types are written to transient data queues, to the VSAM message log files, or to both. See "Task 6: Creating the System Configuration Table (SCT)" for details.
CallPath/CICS does not provide facilities for viewing, printing, or archiving data written to transient data queues. The rest of this section refers only to the facilities provides by CallPath/CICS for writing messages to the VSAM message logs.
All system error and information messages generated by the components of CallPath/CICS are logged by the administration subsystem.
You can display or print all the system messages logged within the time period you specify. Messages are printed in the sequence they were logged, with the oldest message first.
If you choose to display messages, you can specify the order of the messages to be either latest message first or oldest message first. You can also display only those messages associated with a single switch or application program of your choice.
Appendix A, "System Messages" contains a list of messages in numerical order. Each message is explained and you are advised what action to take.
To display messages from the current message log:
Figure 14. System Messages Panel
EQZMK01 System Messages
Type the following information. Then press Enter.
Output destination . . . . . . . . 1 1. Terminal
2. Printer
Start date . . . . . . . . . . . . 1999/04/16
Start time . . . . . . . . . . . . 08:34:32
End date . . . . . . . . . . . . . 1999/04/16
End time . . . . . . . . . . . . . 12:16:03
The following criteria is only applicable if you have selected Terminal as the
output destination:
Message listing sequence . . . . . 1 1. Latest first
2. Oldest first
Switch name . . . . . . . . . . . ________
Program name . . . . . . . . . . . ________
F1=Help F3=Exit F5=Refresh F11=Panel ID F12=Cancel
|
The default start date and time corresponds to the first message in the log. The default end date and time is the current date and time. You can type over these values.
Figure 15. Message Listing Panel
EQZMK02 Message Listing
More: +
Date Time Module Task Program Switch
1999/04/16 08:34:32 EQZ3SUBS 184
EQZ0333I MESSAGE LOG FILE SWAPPED
1999/04/16 08:34:56 EQZ3SUBS 184
EQZ0301I SUBSYSTEM STARTING
1999/04/16 08:35:02 EQZ3SUBS 184
EQZ0325W SWAP LOG REQUEST REJECTED. LOG ALREADY IN USE.
1999/04/16 08:35:11 EQZ2REQH 218
EQZ0516I REQUEST HANDLER STARTING
1999/04/16 08:35:12 EQZ2REQH 218 SWITCH01
EQZ0198I START COMMUNICATIONS REQUEST RECEIVED
1999/04/16 08:35:12 EQZ2REQH 218 SWITCH01
EQZ0506I ESTABLISHING A CONVERSATION WITH A SWITCHSERVER/2
F1=Help F2=Display data F3=Exit F7=Backward F8=Forward
F11=Panel ID F12=Cancel
|
You see the messages corresponding to the search criteria you specified. Each message contains:
You see the Message Data Listing panel.
Figure 16. Message Data Listing Panel
EQZMK03 Message Data Listing Date Time Module Task Program Switch 1999/04/16 08:35:02 EQZ3SUBS 184 EQZ0325W SWAP LOG REQUEST REJECTED. LOG ALREADY IN USE. EQZFLOGA CDECDDCC 58963671 |
For each message, you see the information displayed on the message listing panel, and any associated data. The associated data consists of one or more parameters displayed in both hexadecimal and character format. The parameters associated with each message are described in Appendix A, "System Messages".
You see the System Messages panel (Figure 14).
The default dates and times correspond to the first and last messages in the log. You can overtype these values.
Leave the Switch name and Application name fields blank.
Figure 17 shows an example of a printed message.
Figure 17. Example of a Printed Message
000000 F1F9F9F9 F0F1F1F4 F1F1F2F9 F1F94BF1 F1F00000 000000AA 00000000 40C5D8E9 *19990114112919.110...........EQZ*
000020 F3E2E4C2 E240F1F4 61F0F161 F9F94040 40404040 40F1F17A F2F97AF1 F9404040 *3SUBS 14/01/99 11:29:19 *
000040 40404040 40404040 40404040 404040F1 F5F64040 40404040 40404040 C5D8E9F0 * 156 *
000060 F3F3F3C9 40D4C5E2 E2C1C7C5 40D3D6C7 40C6C9D3 C540E2E6 C1D7D7C5 C4404040 *3331 MESSAGE LOG FILE SWAPPED *
000080 40404040 40404040 40404040 40404040 40404040 40404040 40404040 40404040 * *
0000A0 40404040 40404040 4040
A printed message contains the same information you see when it is displayed on the screen. The message text is followed by parameters. Messages are printed in the sequence they were logged, with the oldest message printed first.
The administration subsystem logs trace records from CallPath/CICS components.
Two types of tracing are available:
During normal operation, switch off all tracing to avoid logging messages you don't need and slowing down your system. When a problem occurs, your service representative will advise you if you need to switch on diagnostic tracing.
You can switch performance tracing and diagnostic tracing on and off independently. You can also start and stop tracing for each module independently, so that you need only switch on tracing for those modules that most interest you. For a given module, diagnostic tracing automatically switches off thirty minutes after you switch it on.
Note: When you use tracing, make sure the CICS data set DFHINTRA is large enough for the trace records or can expand into extents, otherwise the trace records might fill it, causing problems with CICS.
When performance tracing is on, the request handler and event handler each generate two trace records for every message they handle. One record is generated when a message is received for processing, and another when processing is complete and the message is sent. The information contained in each trace record is described in Appendix C, "Log Record Formats". It includes a timestamp and a message type identifier you can use to see how quickly CallPath/CICS processes messages.
Select Performance trace control on the System Administration Menu.
You see the Performance Trace Control panel.
Enter slash (/) next to the request handler or an event handler.
Do one of the following:
Figure 18. Diagnostic Trace Control Panel
EQZMF01 Diagnostic Trace Control
Type / against the modules for which you require diagnostic tracing.
Then press Enter.
More: - +
Module Description
_ EQZ1INIT Initialize Call Profile
_ EQZ1INVK Invoke Feature
_ EQZ1MAKE Make Call
_ EQZ1MON Monitor
_ EQZ1MONS Monitor System Messages
_ EQZ1QPS Query Party Status
_ EQZ1RCV Receive
_ EQZ1RED Redirect Call
_ EQZ1REG Register Ownership
_ EQZ1REJ Reject Call
Total number of modules with trace switched on . . . . . . . . . . : 0
F1=Help F3=Exit F5=Refresh F7=Backward F8=Forward F11=Panel ID
F12=Cancel
|
You see the number of modules for which the trace is active.
Do one of the following:
Note: Tracing switches off automatically thirty minutes after you switch it on. When this happens a message is written to the message log.
This section outlines how you can use the Diagnostic Aids suite of programs to view the contents of some of the CallPath/CICS internal tables. The diagnostic aids will assist you in the development of application systems interfacing with CallPath/CICS, and in problem determination.
The Diagnostic Aids suite provides displays of the Application Name Table, Call Profile Table, and Application Message Queue, and the predefined table sizes on the System Configuration Table.
You can access the tables in one of two ways:
| Note: | You can also access the Call Profile and Application Message Queue directly from the Application Name Table displays. |
Whichever route you use to access a table, PF3 always redisplays the menu.
Figure 19. Diagnostic Aids Menu
EQZMS00 Diagnostic Aids Menu
Select one of the following options by typing its number.
Then press Enter.
_ 1. Program Names
2. Call Profiles
3. Application Message Queue
4. Table Size Definitions
IBM CallPath for OS/390
5655-B34 (C) Copyright by IBM Corp. 1991, 1999. All rights reserved.
Licensed Materials - Property of IBM.
US Government Users Restricted Rights - Use, duplication or disclosure
restricted by GSA ADP Schedule Contract with IBM Corp.
IBM is a registered trademark of International Business Machines Corporation
F1=Help F3=Exit F11=Panel ID F12=Cancel
|
You can access the Application Name Table in two ways:
The screen shown in Figure 20 is displayed.
Figure 20. Program Name Selection
EQZMS02 Program Name Selection Type Program_name to display individual record or leave blank for list of currently assigned names. Then press enter. Program_name . . . . . . . ________ F1=Help F3=Exit F11=Panel ID F12=Cancel |
From the Program Name Selection panel:
Figure 21. List of Program Names
EQZMS06 List of Program Names
Select one of the following options and type / next to the required
program_name. Then press Enter.
_ 1. Program_name details
2. Call_Profile list
3. Application message queue list
More: - +
Program Application Created Last Used Messages
name identifier Date Time Date Time queued
_ PROG1 EQZ001 10/04/99 12:40 10/04/99 15:17 10
_ PROG2 EQZ005 10/04/99 09:16 10/04/99 16:32 105
_ PROG3 EQZ011 10/04/99 11:34 10/04/99 16:47 0
_ PROG4 EQZ006 10/04/99 15:40 10/04/99 17:45 1
_ PROG5 EQZ009 10/04/99 08:23 10/04/99 10:36 7
_ PROG6 EQZ031 10/04/99 10:54 10/04/99 12:28 2
_ PROG7 EQZ015 10/04/99 13:16 10/04/99 18:39 0
_ PROG8765 EQZ007 10/04/99 11:03 10/04/99 13:43 5
F1=Help F3=Exit F5=Refresh F7=Backward F8=Forward F11=Panel ID
F12=Cancel
|
From the List of Program Names List, you can:
Figure 22 shows an individual Application Name Table record. PF keys provide a direct path to other tables:
Figure 22. Program Name Record
EQZMS07 Program Name Record Program_name . . . . . . . . . . . . . . : PROGNAME Application ID . . . . . . . . . . . . . : EQZ001 Date created . . . . . . . . . . . . . . : 10/04/99 Time created . . . . . . . . . . . . . . : 15:36 Date last used . . . . . . . . . . . . . : 10/04/99 Time last used . . . . . . . . . . . . . : 16:27 Number of messages queued . . . . . . . . : 2 Number of call profiles . . . . . . . . . : 4 Receive outstanding . . . . . . . . . . . : YES Message queue overflow . . . . . . . . . : NO Monitoring system messages . . . . . . . : YES F1=Help F3=Exit F4=Messages F5=Refresh F6=Call Profiles F11=Panel ID F12=Cancel |
You can access the Call Profile Table in the following ways:
Figure 23. Call Profile Selection
EQZMS03 Call Profile Selection Type a program_name to list all call profiles belonging to that program, or type a call_profile_ID to display an individual call profile record. Then press Enter. Program_name . . . . . . . ________ (to list all call profiles) Call_profile_ID. . . . . . ________ (to display individual record) F1=Help F3=Exit F11=Panel ID F12=Cancel |
From the Call Profile Selection panel:
If a Program Name is entered, a list of Call Profiles for the Program will be displayed as in Figure 24.
Figure 24. List of Call Profiles
EQZMS08 List of Call Profiles
Program_name . . . . . : PROGNAME
For full call_profile data, type / next to the record you require.
Then press Enter.
More: - +
Call Profile Switch Switch Failure
_ EQZ00101 SOCB1 NO
_ EQZ00102 SOCB1 NO
_ EQZ00103 SOCB22 YES
_ EQZ00105 SWITCHA NO
_ EQZ00106 SOCB1 NO
_ EQZ00107 SOCB22 YES
_ EQZ00108 SWITCHA NO
_ EQZ00109 SOCB1 NO
_ EQZ00110 SOCB1 NO
_ EQZ00111 SOCB1 NO
F1=Help F3=Exit F5=Refresh F7=Backward F8=Forward F11=Panel ID
F12=Cancel
|
From the Call Profile Table display, you can type / next to the required Call Profile to display details about an individual Call Profile Table (see Figure 25).
Figure 25. Call_profile_ID Characteristics
EQZMS09 Call_profile_ID Characteristics
Call_profile_ID . . . . . . . . : EQZ00201
Switch_ID . . . . . . . . . . .: SOCB1
Communications failure . . . .: NO
billed_party . . . . . . . . . .: 01703816985
held_call_recording. . . . . . .: BACH
called_party_alerting_time . . .: 25
More +
Characteristic Setting Value
add_party_direction STL_AP_DIRECTION_SEND_AND_RCV 0
automatic_answer STL_NO_ANSWER 1
call_type STL_VOICE 0
disconnect_execution_option STL_DSC_IF_ROUTING_OR_ALERTING 5
extend_purpose STL_EXTEND_UNSPECIFIED 0
held_call_connection STL_CONNECT_TO_SWITCH_DEFAULT 0
holding_party_callback STL_SWITCH_DEFAULT_CALLBACK 2
holding_party_calling STL_SWITCH_DEFAULT_CALLING 2
F1=Help F3=Exit F5=Refresh F7=Backward F8=Forward F11=Panel ID
F12=Cancel
|
You can access the Application Name Table in the following ways:
Figure 26. Message Queue Selection
EQZMS04 Message Queue Selection Type Program_name. Then press enter. Program_name . . . . . . ________ F1=Help F3=Exit F11=Panel ID F12=Cancel |
A list of the messages on the application's message queue is displayed, as shown in Figure 27
EQZMS05 List of Messages
Program_name . . . . . : PROGNAME
More: - +
Message Type Switch Length Version
Call Alerting SOCB1 1476 0
Call Connected SOCB1 388 0
Call Held SOCB1 904 0
Call Rejected SOCB2 620 0
Call Routed SOCB1 4976 0
Call Transferred SOCB1 648 0
Disconnected SOCB1 2248 0
Feature Invoked SOCB1 108 0
Request Instruction SOCB1 5126 0
Party Status SOCB1 168 0
Response SOCB1 80 0
Switch Status SOCB2 44 0
F1=Help F3=Exit F5=Refresh F7=Backward F8=Forward F11=Panel ID
F12=Cancel
|
You can access the System Configuration Table in one of two ways:
A screen showing defined table sizes and selected System Configuration Table fields is displayed (see Figure 28).
Figure 28. System Configuration Table
EQZMS01 System Configuration Table Display
Date format . . . . . . . . . . . . . . . : 04/10/99 (DD/MM/YY)
Time format . . . . . . . . . . . . . . . : 13:56:27
Maximum storage limit (Kbytes) . . . . . . : 1024
Current storage (Kbytes) . . . . . . . . . : 337
Application message queue capacity . . . . : 100
Application time limit (minutes) . . . . . : 0
Table Primary space Secondary space Secondary extents
(Kbytes) (Kbytes)
EQSWITCH 5 2 0
EQREQ 7 2 1
EQFONE 5 2 0
EQMON 5 2 0
EQANT 75 0 2
EQAMQ 180 60 0
EQCALLPR 50 22 0
EQREG 5 2 0
EQTASK 5 2 0
F1=Help F3=Exit F5=Refresh F11=Panel ID F12=Cancel
|
For each switch, the request handler counts the number of each type of message it sends, and the event handler counts the number of each type of message it receives. Every fifteen minutes, both the request handler and the event handler send the information to the administration subsystem and reset their counters to zero.
If tracing is switched on in CallPath SwitchServer/2, it also counts messages and sends the counts every fifteen minutes to CallPath/CICS for logging. CallPath SwitchServer/2 counts the number of messages received from, and sent to, all host computers for which it has tracing switched on. It also counts the number of messages received from, and sent to, the switch. The administration subsystem writes all the statistics to the traffic log file.
Notes:
Using the panels described here you can display the message traffic statistics collected by the administration subsystem since it was last started.
To display message traffic statistics:
Figure 29. Message Traffic Statistics Panel
EQZMD01 Message Traffic Statistics
Select the type of traffic statistics you require.
Type of traffic . . . . . . _ 1. Outgoing messages from host
2. Incoming messages to host
3. SwitchServer/2 messages between host & switch
Type / next to the switch you require. Then press Enter.
. SWITCH01
. SWITCH02
F1=Help F3=Exit F5=Refresh F7=Backward F8=Forward F11=Panel ID
F12=Cancel
|
Figure 30. Traffic Statistics Display Panel
EQZMD02 Traffic Statistics Display
Switch name . . . . . : SWITCH01
Traffic type . . . . . : 1
Start date . . . . . . : 16/04/1999
Start time . . . . . . : 08:50:31 More: +
<------- Last 4 counts -------> Session
Description 10:05 10:20 10:35 10:50 Total
Add Party 3 0 1 0 10
Alternate Call 0 2 1 0 7
Answer Call 44 75 61 34 353
Conference Call 4 11 3 7 42
Disconnect 9 2 4 1 25
Extend Call 0 1 4 2 9
Hold Call 12 7 9 7 59
Invoke Feature 0 0 2 0 3
Immediate Transfer 20 5 9 13 94
Make Call 35 51 46 49 287
F1=Help F3=Exit F5=Refresh F7=Backward F8=Forward F11=Panel ID
F12=Cancel
|
You see statistics for the switch and the type of traffic that you chose. The Start date and Start time show when the administration subsystem was started. In Figure 30, the traffic type is 1, which refers to outgoing messages from the host.
You can see that the Add_Party program call was sent by the host three times during the 15-minute period ending at 10:05 but was not sent at all during the next 15-minute period ending at 10:20. The Session Total shows the total usage of each message since the administration subsystem was started.
The messages included in each type of traffic are shown in Table 15.
Table 15. Messages Included in Each Type of Traffic
| Outgoing Messages from Host | Incoming Messages to Host | SwitchServer/2 Counts |
|---|---|---|
| Add_Party | Call_Alerting | Host to SwitchServer/2 |
| Alternate_Call | Call_Alternated | SwitchServer/2 to host |
| Answer_Call | Call_Conferenced | SwitchServer/2 to switch |
| Conference_Call | Call_Connected | Switch to SwitchServer/2 |
| Disconnect | Call_Held | |
| Extend_Call | Call_Parked | |
| Hold_Call | Call_Picked | |
| Immediate_Transfer | Call_Rejected | |
| Invoke_Feature | Call_Routed | |
| Make_Call | Call_Transferred | |
| Monitor | Data_Collected | |
| Query_Party_Status | Feature_Invoked | |
| Redirect_Call | Disconnected | |
| Reject_Call | Network_Reached | |
| Release_Program_Name | Party_Status | |
| Retrieve_Call | Request_Instruction | |
| Return_Control | Response | |
| Transfer_Call | Setup | |
| Trigger | Switch_Status | |
| Send_Device_Data | Session_Started | |
| Session_Start | Session_Ended | |
| Session_End
| Switch_Information
|
|
+-------------------Product-sensitive programming interface--------------------+
When an application program uses CallPath/CICS, CallPath/CICS reserves resources for the application to use for its message queues. If the application program makes an Identify_Program_Name (STLIDEN) program call, CallPath/CICS associates resources with the application program's program name. You can limit the storage used by each Identify_Program_Name program call using the MAXMSGS parameter of the CallPath/CICS System Configuration Table (SCT).
Normally, when an application program has finished using CallPath/CICS, it issues a Release_Program_Name program call (STLREL) to free the resources. If an application program does not issue this program call, it continues to use the CallPath/CICS resources even when it is no longer using CallPath/CICS. When this occurs, you can release resources using the Queue Clean program.
The Queue Clean program frees resources reserved for application programs that have not made a program call for longer than a predetermined period of time (up to 48 hours), which is defined in the system configuration table.
To run the Queue Clean program:
A message to the terminal informs you when Queue Clean processing is complete. If a program_name cannot be released, a message is produced on the system log. (If EQQC is started by a user-written application program, the program does not write a message to the terminal when Queue Clean processing is complete.)
+----------------End of Product-sensitive programming interface----------------+
