top - download
⟦a94bdde0e⟧ Wang Wps File
Length: 47823 (0xbacf)
Types: Wang Wps File
Notes: CPS/SDS/009
Names: »1128A «
Derivation
└─⟦fad1c67d1⟧ Bits:30006043 8" Wang WCS floppy, CR 0069A
└─ ⟦this⟧ »1128A «
WangText
0…00……00……00……00…C…02……00……00…C
…10… …0f……0e……0f……05……0e……09……0e……0e……0e…
…0e……06……0e……07……0d……0d……0d……02……0d……06……0c……0a……0c……0e……0c……02……0c……07……0b……0c……0b……0f……0b…
…0a……08……0a……0b……0a……0c……0a……00……0a……05……0a……06……09……0c……09……86…1 …02… …02… …02…
…02…CPS/SDS/009
…02…HUH/810801…02……02…
LOGGING
…02……02…CAMPS
…06…1 …02… …02… …02… …02… …02… …02… …02… …02… …02… …02… …02… …02… …02…
T̲A̲B̲L̲E̲ ̲O̲F̲ ̲C̲O̲N̲T̲E̲N̲T̲S̲
1 GENERAL ..........................................
6
1.1 PURPOSE AND SCOPE ..............................
6
1.2 APPLICABLE DOCUMENTS AND PROJECT REFERENCES ....
6
1.2.1 Applicable Documents .......................
7
1.2.2 Project References .........................
7
1.3 TERMS AND ABBREVIATIONS ........................
7
1.3.1 Terms .....................................
7
1.3.2 Abbreviations .............................
7
2 SUMMARY OF REQUIREMENTS ..........................
8
2.1 PACKAGE DESCRIPTION ............................
8
2.2 PACKAGE FUNCTIONS ..............................
11
2.2.1 Main Functions at Normal Operation .........
11
2.2.2 Functional Responsibilities ................
11
2.2.2.1 Initialization, Close Down and Restart
. 11
2.2.2.1.1 Initialize and Restart .............
12
2.2.2.1.1.1 Start Up Types ................
12
2.2.2.1.1.2 Log Package Initialize ........
13
2.2.2.1.2 Close Down .........................
14
2.2.2.2 Checkpointing and Recovery .............
14
2.2.2.2.1 Checkpointing ......................
14
2.2.2.2.2 Recovery ...........................
15
2.2.2.2.2.1 Warm 2 ........................
15
2.2.2.2.2.2 Warm 3 ........................
15
2.2.2.3 Error Detecting and Error Handling .....
16
2.2.2.4 Integrity of Operation .................
17
2.2.2.5 Data Collection ........................
17
2.2.2.6 Security ...............................
17
2.3 CHARACTERISTICS ................................
17
2.3.1 Timing .....................................
17
2.3.1.1 Collect Run Time .......................
18
2.3.1.2 Trace Run Time .........................
18
2.3.2 Throughput .................................
18
2.3.3 Flexibility ................................
19
2.3.4 Accuracy ...................................
19
3 ENVIRONMENT .....................................
19
3.1 EQUIPMENT ......................................
19
3.2 SOFTWARE .......................................
19
3.2.1 System Software ............................
19
3.2.2 Development Support Software ..............
20
3.3 INTERFACES .....................................
20
3.3.1 External Interfaces ........................
20
3.3.2 Package Interfaces .........................
20
3.3.2.1 Application Package Interface ..........
20
3.3.2.2 SSC Interface ..........................
21
3.3.2.3 TEP Interface ..........................
21
3.3.2.4 CSF Interface ..........................
21
3.3.2.5 MMS Interface ..........................
22
3.3.2.6 SAR Interface ..........................
22
3.4 FUNCTIONS MAINTAINED BY OTHER PACKAGES .........
24
4 PACKAGE DESIGN ...................................
25
4.1 PACKAGE OVERVIEW ...............................
25
4.1.1 Functional Specification ...................
27
4.1.1.1 Functional Breakdown ...................
27
4.1.1.1.1 Log Collection Functions ...........
27
4.1.1.1.2 Log Trace Functions ................
28
4.1.2 Software Specification .....................
30
4.1.3 Data Flow and Control Logic ................
30
4.1.4 Package Data ...............................
34
4.1.5 Common Data ................................
34
4.1.5.1 Log Record Data ........................
34
4.1.5.1.1 Log Record Parameters ..............
34
4.1.5.1.2 Log Record Type ....................
37
4.1.5.2 Log Trace Parameters ...................
44
4.1.6 Interfaces .................................
44
4.1.6.1 External Interfaces ....................
44
4.1.6.2 Package Interfaces .....................
45
4.1.6.2.1 Applications Interfaces ............
45
4.1.6.2.2 Interface to SSC ...................
48
4.1.6.2.3 Interface to TEP ...................
48
4.1.6.2.4 Interface to CSF and SFM ...........
48
4.1.6.2.5 Interface to SAR ...................
49
4.1.6.3 Sub-Package Interfaces .................
49
4.2 Sub-Package Specification ......................
49
4.2.1 Log Collect Sub-Package Specification ......
49
4.2.1.1 Functional Specification ...............
50
4.2.1.1.1 Log Record Collection ..............
50
4.2.1.1.2 Log Record Checkpointing ..........
50
4.2.1.1.3 Log Record Printout ................
51
4.2.1.1.4 Log Record Store ...................
51
4.2.1.2 Software Structure .....................
53
4.2.1.3 Data Flow and Control Logic ............
55
4.2.1.3.1 Control Logic ......................
55
4.2.1.3.1.1 Coroutine Interrelationship ....
55
4.2.1.3.2 Data Flow ..........................
58
4.2.1.4 Sub-Package Data .......................
70
4.2.1.5 Sub-Package Interface ..................
71
4.2.2 Log Trace Sub-Package Specification ........
71
4.2.2.1 Functional Specification ...............
71
4.2.2.1.1 Receive and Validate Trace Keys ....
71
4.2.2.1.2 Retrieve Pertinent Log CIFs ........
72
4.2.2.1.3 Trace Log Records ..................
72
4.2.2.1.4 Return Traced Log Records ..........
72
4.2.2.2 Software Structure .....................
74
4.2.2.3 Data Flow and Control Logic ............
76
4.2.2.3.1 Data Flow ..........................
76
4.2.2.4 Sub-Package Data .......................
84
4.2.2.5 Sub-Package Interface ..................
84
4.3 MEMORY LAY OUT .................................
85
1̲ ̲ ̲G̲E̲N̲E̲R̲A̲L̲
1.1 P̲U̲R̲P̲O̲S̲E̲ ̲A̲N̲D̲ ̲S̲C̲O̲P̲E̲
The package specification for logging (LOG no. 4040)
is written to fulfil the following objectives:
1) To provide detailed definition of the package functions
and software architecture.
2) To provide user operational and development personnel
details of the ongoing analysis.
3) To define in detail the interfaces with other packages
and to describe their facilities.
The design specification is a software description
of the LOG package, down to a level where all modules
are defined.
Higher level design specifications can be found in
the system design document.
The application interfaces to LOG are described in
this document.
All interfaces and data are described in detail in:
- CPS/ICD/009 Software Interface Control Document
- CPS/DBD/001 Database Design Document
1.2 A̲P̲P̲L̲I̲C̲A̲B̲L̲E̲ ̲D̲O̲C̲U̲M̲E̲N̲T̲S̲ ̲A̲N̲D̲ ̲P̲R̲O̲J̲E̲C̲T̲ ̲R̲E̲F̲E̲R̲E̲N̲C̲E̲S̲
1.2.1 A̲p̲p̲l̲i̲c̲a̲b̲l̲e̲ ̲D̲o̲c̲u̲m̲e̲n̲t̲s̲
- CPS/210/SYS/0001 CAMPS System Requirements
- CPS/SDS/001 CAMPS System Design
- CPS/SDS/002 CAMPS System Function
- CPS/SDS/004 System Status and Control
- CPS/SDS/007 Storage and Retrieval
- CPS/SDS/010 Traffic Handling
- CPS/SDS/012 Terminal Package
1.2.2 P̲r̲o̲j̲e̲c̲t̲ ̲R̲e̲f̲e̲r̲e̲n̲c̲e̲s̲
N/A.
1.3 T̲E̲R̲M̲S̲ ̲A̲N̲D̲ ̲A̲B̲B̲R̲E̲V̲I̲A̲T̲I̲O̲N̲S̲
1.3.1 T̲e̲r̲m̲s̲
Applications - packages such as TEP which contribute
with log information to the LOG package.
Diag.-id. - Diagram identification. Identify the number
of the diagram and explain the diagram page out of
the total number.
1.3.2 A̲b̲b̲r̲e̲v̲i̲a̲t̲i̲o̲n̲s̲
CIF CAMPS Information Field (item)
CSF CAMPS System Function
SAR Storage and Retrieval
MMS Storage and File Management
SSC System Status and Control
TEP Terminal Package
MMS Message Management System within MMS
MDCO Message Distribution Control Operator
MSO Message Service Operator
TOS Terminal Operating System
DAMOS CR80D Advanced Multiprocessor Operating System
PD Preliminary Design
SP Sub-Package
LOG Logging Package
2̲ ̲ ̲S̲U̲M̲M̲A̲R̲Y̲ ̲O̲F̲ ̲R̲E̲Q̲U̲I̲R̲E̲M̲E̲N̲T̲S̲
2.1 P̲A̲C̲K̲A̲G̲E̲ ̲D̲E̲S̲C̲R̲I̲P̲T̲I̲O̲N̲
The tasks of the LOG package are:
a) Collect log records from other applications.
b) Send the incoming log information to SSC for checkpointing.
c) Send the incoming log information to TEP for printout.
d) Store the log information by means of MMS and SAR
when five log records are collected or when 30
seconds have elapsed.
e) Recover the checkpointed log records after system
break down.
f) Trace log records requested by TEP using SAR to
retrieve the log CIF-ref. in question.
A chart with the LOG package and its inter-relationships
with other packages is shown on fig. 2.1-1.
Explanation to the inter-relationships chart fig. 2.1-1:
Step 1: The application transfers the log records to
LOG.
Step 2: When a specific log record is safely stored,
an acknowledge is returned.
Step 3: The incoming log records are transferred to
SSC for checkpointing.
Step 4: Log record checkpoint acknowledge returned
to LOG.
Step 5: The incoming log records are transferred to
the TEP supervisor log print queue.
Step 6: Request to MMS for creation of log CIF.
Step 7: Return of the created log CIF-ref.
Step 8: Append log record to MMS.
Step 9: Set-up time monitor in CSF.
Step 10: Receive time out from time monitor.
Step 11: Request SAR to store the log CIF.
Step 12: TEP request LOG to retrieve log records.
Step 13: LOG request SAR to find a specified log
CIF-ref.
Step 14: Return of requested log CIF-ref.
Step 15: Retrieve log records from disk.
Step 16: Return the traced log records.
Fig. 2.1-1…01…LOG Inter-Relationships.
2.2 P̲A̲C̲K̲A̲G̲E̲ ̲F̲U̲N̲C̲T̲I̲O̲N̲S̲
2.2.1 M̲a̲i̲n̲ ̲F̲u̲n̲c̲t̲i̲o̲n̲s̲ ̲a̲t̲ ̲N̲o̲r̲m̲a̲l̲ ̲O̲p̲e̲r̲a̲t̲i̲o̲n̲
The main functions at normal operation are:
a) Collection of log records from TEP and THP.
TEP and THP transfer the log records to the LOG
package via buffers and queue elements.
The log records are collected in main memory.
b) Append the collected log records.
Every 30 seconds or when 5 log records are collected,
the collected log records are appended to the log
CIF.
Every 10 minutes the present used log CIF-ref.
are stored by SAR and a new requested.
Log records are stored online for a period of at
least 24 hours and offline for at least 30 days.
c) Checkpointing of log records.
The incoming log records are checkpointed by SSC
in the standby PU.
d) Log record printout.
The incoming log records are queued in the supervisor
log print queue.
e) Trace of stored log information.
The stored log information is retrieved and traced
on supervisor request. The trace command and trace
keys are sent to LOG via the TEP package.
The requested log records are appended in a trace
CIF and the trace CIF-ref. is queued in the supervisor
print queue.
2.2.2 F̲u̲n̲c̲t̲i̲o̲n̲a̲l̲ ̲R̲e̲s̲p̲o̲n̲s̲i̲b̲i̲l̲i̲t̲i̲e̲s̲
2.2.2.1 I̲n̲i̲t̲i̲a̲l̲i̲z̲a̲t̲i̲o̲n̲,̲ ̲C̲l̲o̲s̲e̲ ̲D̲o̲w̲n̲ ̲a̲n̲d̲ ̲R̲e̲s̲t̲a̲r̲t̲
2.2.2.1.1 I̲n̲i̲t̲i̲a̲l̲i̲z̲e̲ ̲a̲n̲d̲ ̲R̲e̲s̲t̲a̲r̲t̲
The initialize and restart software is placed at the
LOG package start up entry point.
2.2.2.1.1.1 S̲t̲a̲r̲t̲-̲u̲p̲ ̲T̲y̲p̲e̲s̲
D̲e̲a̲d̲ ̲1̲,̲ ̲D̲e̲a̲d̲ ̲2̲ ̲a̲n̲d̲ ̲W̲a̲r̲m̲ ̲3 - Used after ordered close
down and used as system initialize.
These start up types are all treated as Dead 1. No
recovery actions are taken and all the used data area
will be initialized.
C̲o̲l̲d̲
Cold is treated like Dead 1 apart from clearing all
online log CIFs.
W̲a̲r̲m̲ ̲1 - Used after total system error.
Warm 1 is treated like Dead 1 apart from recovering
the last used CIF-ref.
W̲a̲r̲m̲ ̲2 - Used after switch-over.
Warm 2 is treated like Warm 1 apart from recovering
of the checkpointed log records.
I̲n̲s̲e̲r̲t̲i̲o̲n̲ ̲o̲f̲ ̲S̲t̲a̲n̲d̲ ̲b̲y̲ ̲P̲U̲
When the stand-by PU is inserted, all logging activities
will be normal 30 sec. after the initialize command
from SSC is executed.
For detailed description please refer CPS/SDS/004,
section 2.2.2.1.1.1.
2.2.2.1.1.2 L̲o̲g̲ ̲P̲a̲c̲k̲a̲g̲e̲ ̲I̲n̲i̲t̲i̲a̲l̲i̲z̲e̲
The initialize command from SSC will cause the following
actions:
a) Identify command and call INIT routine
b) If the command is COLD, then CLEAR ̲CIF is called.
c) Send completion code to SSC.
T̲h̲e̲ ̲I̲N̲I̲T̲ ̲r̲o̲u̲t̲i̲n̲e̲:
1) Read queue element from SSC and get init. file-ID
2) Read init. file and load initialize parameters
into log data area: Record validation table,
log trace table.
3) Request 10 min. and 30 sec. time out.
4) Initialize internal log date area.
5) Initialize flags.
6) Initialize semaphores.
7) Initialize coroutines
8) Create and checkpoint log CIF
9) Return to calling routine
T̲h̲e̲ ̲C̲L̲E̲A̲R̲ ̲C̲I̲F̲ ̲r̲o̲u̲t̲i̲n̲e̲:
1) Request online CIF-ref. (SAR)
2) Clear online CIF contents (MMS)
3) Return to calling routine
2.2.2.1.2 C̲l̲o̲s̲e̲ ̲D̲o̲w̲n̲
The close down software is resident in the COLLECT
coroutine and is called COMMAND.
When COLLECT receives a close down command from SSC,
the following actions are taken:
a) Empty incoming log queue.
b) Append the collected log records to the log CIF.
c) Update CIF-index table
d) Store log CIF
e) Send completion code to SSC.
For further description please refer to section 4.2.1.3.2.
2.2.2.2 C̲h̲e̲c̲k̲ ̲P̲o̲i̲n̲t̲i̲n̲g̲ ̲a̲n̲d̲ ̲R̲e̲c̲o̲v̲e̲r̲y̲
2.2.2.2.1 C̲h̲e̲c̲k̲ ̲P̲o̲i̲n̲t̲i̲n̲g̲
The checkpointing of the incoming log records is implemented
by SSC.
Before the log package sends log records for checkpointing,
they are assigned log record serial numbers. The log
record serial number consists of a log record count
(0-4) and a checkpoint serial number.
The log record count indicates the stand-by PU buffer
address. The checkpoint serial number are used by LOG
when the checkpointed log records are recovered.
Max. five log records are checkpointed by SSC in the
standby PU. Earlier checkpointed log records will
be overwritten. The record number tells which one
to overwrite.
The present used log CIF is checkpointed on disk.
2.2.2.2.2 R̲e̲c̲o̲v̲e̲r̲y̲
The recovery routines are called when a recovery command
is received from SSC.
There are two recovery commands: Warm 1 and Warm 2.
2.2.2.2.2.1 W̲a̲r̲m̲ ̲1
A Warm 1 command will cause the following actions:
a) Identify command and call INIT routine.
b) Call REC ̲CIF routine
c) Send completion code to SSC.
T̲h̲e̲ ̲R̲E̲C̲ ̲C̲I̲F̲ ̲r̲o̲u̲t̲i̲n̲e̲
1) Recover the last used CIF-ref. from disk.
2) Return to calling routine.
2.2.2.2.2.2 W̲a̲r̲m̲ ̲2
A Warm 2 command will cause the following actions:
a) Identify command and call INIT routine.
b) Call REC ̲CIF routine
c) Call REC ̲LOG routine
d) Send completion code to SSC.
T̲h̲e̲ ̲R̲E̲C̲ ̲L̲O̲G̲ ̲r̲o̲u̲t̲i̲n̲e̲
1) Receive the five checkpointed log records from
SSC.
2) Recover the proper log records in the log collect
data area.
3) Set log record serial number.
4) Return to calling routine.
2.2.2.3 E̲r̲r̲o̲r̲ ̲D̲e̲t̲e̲c̲t̲i̲n̲g̲ ̲a̲n̲d̲ ̲E̲r̲r̲o̲r̲ ̲H̲a̲n̲d̲l̲i̲n̲g̲
Errors detected by LOG are:
a) Error in the incoming log records. (length, type)
Checkable log parameters are validated for upper
and lower boundaries.
b) No buffer or CIF available from MMS.
c) No queue element available from CSF.
d) Error in trace commands and trace keys from TEP.
E̲r̲r̲o̲r̲ ̲H̲a̲n̲d̲l̲i̲n̲g̲
All errors are reported to SSC.
Error a: The incoming log record is deleted and
a completion code is returned to the sender.
Error b
and c: The collect action is stopped and dummy
mode entered.
Error d: The command is deleted and a completion
code returned to TEP.
2.2.2.4 I̲n̲t̲e̲g̲r̲i̲t̲y̲ ̲o̲f̲ ̲O̲p̲e̲r̲a̲t̲i̲o̲n̲
To avoid errors in the log collection mechanism, all
the incoming log records are validated.
The checkable log parameters are checked for upper
and lower boundaries. The log record type and length
is also checked.
2.2.2.5 D̲a̲t̲a̲ ̲C̲o̲l̲l̲e̲c̲t̲i̲o̲n̲
The collection of log records is done by means of queue
elements and the buffer functions implemented by CSF.
The log records are transferred to the incoming log
queue by sending a queue element with a reference to
a log record buffer.
When a log record is received by LOG, an acknowledge
queue element is returned to the log record sender
when the log record is safety stored.
2.2.2.6 S̲e̲c̲u̲r̲i̲t̲y̲
No security action is performed by LOG.
2.3 C̲H̲A̲R̲A̲C̲T̲E̲R̲I̲S̲T̲I̲C̲S̲
2.3.1 T̲i̲m̲i̲n̲g̲
The incoming log records will be stored on disk when
30 seconds have elapsed or 5 records are collected.
Every 10 minutes, the presently used CIF will be closed
and stored. This action is only taken if the CIF contains
appended log records.
Log records are stored online for a period of at least
24 hours.
Log records are stored offline for a period of at least
30 days.
2.3.1.1 C̲o̲l̲l̲e̲c̲t̲ ̲R̲u̲n̲ ̲T̲i̲m̲e̲
The CPU time used by LOG during the log collect procedure:
a) For each log record, collected, va-
lidated, checkpointed and queued
for print out : 1
ms
b) Appending of five collected
log records : 0.2
ms
c) Storage of old log CIF and re
quest of new log CIF every 10 min. : 0.1
ms
Estimated average CPU time used by
LOG for each received log record : 1.1
ms
Estimated CPU time used by LOG in
a busy min. : 160
ms
2.3.1.2 T̲r̲a̲c̲e̲ ̲R̲u̲n̲ ̲T̲i̲m̲e̲
The average CPU time used by log,
tracing a ten busy minute log CIF
is estimated to : 100
ms
The trace procedure is executed in such a way that
no significant performance reduction occurs.
2.3.2 T̲h̲r̲o̲u̲g̲h̲p̲u̲t̲
The amount of log records are estimated to be 145 log
records per busy minute, 7000 log records per busy
hour and 35000 per busy 24 hours.
2.3.3 F̲l̲e̲x̲i̲b̲i̲l̲i̲t̲y̲
The modularity of the LOG package facilitates implementation
of changes in the log record format and trace format
as well as changes in the log record collect and store
timing.
2.3.4 A̲c̲c̲u̲r̲a̲c̲y̲
Not applicable to LOG.
3̲ ̲ ̲E̲N̲V̲I̲R̲O̲N̲M̲E̲N̲T̲S̲
3.1 E̲Q̲U̲I̲P̲M̲E̲N̲T̲
The software within the LOG package is online resident
in one processing unit.
Main memory data area consists of buffer and data area.
Main memory buffer area size will be 1000 words.
Main memory data area size will be 900 words.
3.2 S̲O̲F̲T̲W̲A̲R̲E̲
3.2.1 S̲y̲s̲t̲e̲m̲ ̲S̲o̲f̲t̲w̲a̲r̲e̲
The system software environment to LOG are:
- CAMPS System Function
- Message Management System
3.2.2 D̲e̲v̲e̲l̲o̲p̲m̲e̲n̲t̲ ̲S̲u̲p̲p̲o̲r̲t̲ ̲S̲o̲f̲t̲w̲a̲r̲e̲
Development support software is standard DAMOS and
TOS resident in a single CR 80D.
3.3 I̲N̲T̲E̲R̲F̲A̲C̲E̲
3.3.1 E̲x̲t̲e̲r̲n̲a̲l̲ ̲I̲n̲t̲e̲r̲f̲a̲c̲e̲s̲
There are no external interfaces to LOG.
3.3.2 P̲a̲c̲k̲a̲g̲e̲ ̲I̲n̲t̲e̲r̲f̲a̲c̲e̲s̲
An interface overview is shown in figure 3.3.2-1.
3.3.2.1 A̲p̲p̲l̲i̲c̲a̲t̲i̲o̲n̲ ̲P̲a̲c̲k̲a̲g̲e̲ ̲I̲n̲t̲e̲r̲f̲a̲c̲e̲
The LOG package receive log records from Traffic Handling
Package and Terminal Package. This interface is referred
to as application package interface.
The application package interface is implemented by
means of queue elements and the buffer function implemented
by CSF.
The buffers are used to transport the log records from
the applications to LOG.
When a log record is safely stored an acknowledge is
returned to the sender in an answer queue.
3.3.2.2 S̲S̲C̲ ̲I̲n̲t̲e̲r̲f̲a̲c̲e̲
The interface to SSC is implemented by means of queues.
The queues are used to:
a) Transfer commands from SSC
b) Transfer log records to SSC for checkpointing
c) Receive acknowledge from SSC
d) Send completion codes to SSC
3.3.2.3 T̲E̲P̲ ̲I̲n̲t̲e̲r̲f̲a̲c̲e̲
The interface to TEP is implemented by means of queues,
buffers, and a common buffer area. The common buffer
area may contain up to 50 log records waiting for printout.
Incoming log records are stored in a common buffer
area, queue elements with buffer pointers are sent
to the TEP printout queue.
Log trace commands are sent to LOG via a queue. The
queue elements contain pointers to buffers with trace
keys. The traced log records are stored in a CIF and
the CIF-ref. will be returned to TEP in a queue element.
3.3.2.4 C̲S̲F̲ ̲I̲n̲t̲e̲r̲f̲a̲c̲e̲
The interface to CSF is implemented by means of monitor
calls.
The monitor calls are used e.g. when requesting for
a queue element and time out.
Please refer to CPS/ICD/009.
3.3.2.5 M̲M̲S̲ ̲I̲n̲t̲e̲r̲f̲a̲c̲e̲
Each command from an application process to MMS is
implemented as a system call to MMON.
For detailed description please refer to CPS/ICD/009.
3.3.2.6 S̲A̲R̲ ̲I̲n̲t̲e̲r̲f̲a̲c̲e̲
The SAR interface is implemented with queues.
The queues are used to:
a) Request SAR for a CIF-ref. Input is a time interval
and SAR log-indicator.
b) Receive CIF-ref. from SAR.
c) Request SAR for cataloging and storage of a CIF.
Input is a CIF-ref.
Fig. 3.3.2-1…01…LOG Interfaces
Queue names used in fig. 3.3.2-1.
Q 1 IN LOG Incoming log queue
Q 2 APP ANSWQ Application answer queue
Q 3 IN COMM Incoming command queue
Q 4 COMM RESP Command response queue
Q 5 CHKP RESP Checkpoint response queue
Q 6 CTQ Checkpoint transmission
queue
Q 7 30 S TIM 30 seconds time out queue
Q 8 10MTIM 10 minutes time out queue
Q 9 RRQ Retrieval request queue
Q 10 RET RESP Retrieve response queue
Q 11 ISQ Incoming storage queue
Q 12 TRACE COMM Trace command queue
Q 13 SPQ Supervisor print queue
Q 14 SLPQ Supervisor log print queue
3.4 F̲U̲N̲C̲T̲I̲O̲N̲S̲ ̲M̲A̲I̲N̲T̲A̲I̲N̲E̲D̲ ̲B̲Y̲ ̲O̲T̲H̲E̲R̲ ̲P̲A̲C̲K̲A̲G̲E̲S̲
The functions maintained by other packages are:
- Recovery by SSC
- Error handling by SSC
- Check pointing by SSC
4̲ ̲ ̲P̲A̲C̲K̲A̲G̲E̲ ̲D̲E̲S̲I̲G̲N̲
4.1 P̲A̲C̲K̲A̲G̲E̲ ̲O̲V̲E̲R̲V̲I̲E̲W̲
The main events in the LOG package are shown by means
of a diagram fig. 4.1-1 and appertaining text.
It is natural to divide the events in a log collecting
part and a log tracing part.
L̲o̲g̲ ̲C̲o̲l̲l̲e̲c̲t̲i̲o̲n̲ - Log records, received from the applications
are collected.
A̲p̲p̲e̲n̲d̲ ̲l̲o̲g̲ ̲r̲e̲c̲o̲r̲d̲s̲ - The collected log records are
appended to a created CIF when 30 seconds have elapsed
or when 5 records are collected.
S̲t̲o̲r̲e̲ ̲l̲o̲g̲ ̲C̲I̲F̲ - The log CIFs are stored with 10 minute
intervals and new log CIFs are created. The not used
disk area within the CIFs is cut off by MMS upon this
storage.
SAR catalogs the CIF-ref. for later retrieval.
L̲o̲g̲ ̲t̲r̲a̲c̲i̲n̲g̲ - The supervisor's log trace requests are
transferred via TEP to the LOG package. A number of
trace keys tells the log trace mechanisms, which of
the stored log records to retrieve.
R̲e̲t̲r̲i̲e̲v̲a̲l̲ ̲o̲f̲ ̲l̲o̲g̲ ̲C̲I̲F̲s̲ - The retrieved log CIFs are
traced and the wanted log records selected.
A̲p̲p̲e̲n̲d̲ ̲l̲o̲g̲ ̲r̲e̲c̲o̲r̲d̲s̲ - The selected log records are appended
to an especially created CIF.
T̲r̲a̲n̲s̲f̲e̲r̲ ̲l̲o̲g̲ ̲C̲I̲F̲-̲r̲e̲f̲.̲ ̲t̲o̲ ̲T̲E̲P̲ - The log CIF-ref. and
a completion code is returned to TEP.
Fig. 4.1-1…01…Package Events
4.1.1 F̲u̲n̲c̲t̲i̲o̲n̲a̲l̲ ̲S̲p̲e̲c̲i̲f̲i̲c̲a̲t̲i̲o̲n̲
4.1.1.1 F̲u̲n̲c̲t̲i̲o̲n̲a̲l̲ ̲B̲r̲e̲a̲k̲d̲o̲w̲n̲
The functions implemented by LOG are separated in two
groups:
1) Log collection functions
2) Log trace functions
4.1.1.1.1 L̲o̲g̲ ̲C̲o̲l̲l̲e̲c̲t̲i̲o̲n̲ ̲F̲u̲n̲c̲t̲i̲o̲n̲s̲
Log records, generated by the applications, are transferred
to the incoming log queue. The incoming log records
are read by the LOG application and four things happen:
a) The log record length and parameters are checked.
If any errors are found, the record will be returned
to the sender with an error code.
b) The log record is sent to SSC for checkpointing.
Acknowledge is awaited. When received, an acknowledge
is returned to the log record sender.
c) The log record is sent to the TEP printout queue.
If the queue contains more than 49 unprinted log
records, this function will be skipped.
d) The log record is stored in the internal data area
in LOG.
S̲t̲o̲r̲a̲g̲e̲ ̲o̲f̲ ̲L̲o̲g̲ ̲R̲e̲c̲o̲r̲d̲s̲
The following functions are implemented by LOG when
log records are stored:
a) Every 30 sec. or when five log records are collected
in the internal data area, the log records are
appended to a created CIF.
b) Every 10 minutes, the CIF is stored and a new is
generated. If no records were appended to the
CIF, no action is taken.
4.1.1.1.2 L̲o̲g̲ ̲T̲r̲a̲c̲e̲ ̲F̲u̲n̲c̲t̲i̲o̲n̲s̲
The stored log records may be retrieved with respect
to specific trace keys and sent to supervisors position
via TEP. The following functions are implemented by
LOG when log records are retrieved:
a) The trace keys are received from TEP.
b) The pertinent time interval is computed and the
CIF-refs. are retrieved.
c) The wanted log records are traced and appended
to a trace CIF.
d) The trace CIF-ref is sent to TEP via a queue element
when the trace action is terminated.
Fig. 4.1.1.1.2-1…01…Functional Breakdown
4.1.2 S̲o̲f̲t̲w̲a̲r̲e̲ ̲S̲p̲e̲c̲i̲f̲i̲c̲a̲t̲i̲o̲n̲
The LOG package consists of two sub-packages:
1) Log collect sub-package
2) Log trace sub-package
The functions allocated to each of the sub-packages
correspond with the functional break-down in chapter
4.1.1.1.
The two sub-packages are independent of one another
and they have no common data area.
The sub-packages consist of a number of modules. The
software structure for the sub-packages are included
in chapter 4.2.
4.1.3 D̲a̲t̲a̲ ̲F̲l̲o̲w̲ ̲a̲n̲d̲ ̲C̲o̲n̲t̲r̲o̲l̲ ̲L̲o̲g̲i̲c̲
The data flow and control for LOG is presented in HIPO
diagrams.
The HIPO diagrams in this chapter show the data flow
and control overview for the two sub-packages.
E̲x̲p̲l̲a̲n̲a̲t̲i̲o̲n̲ ̲t̲o̲ ̲H̲I̲P̲O̲ ̲s̲u̲b̲-̲t̲e̲x̲t̲
Diag. ID - Diagram reference, number and page number.
Name - The name of the described software module
Descr. - Description of the HIPO contents.
3 HIPOs
HIPO
HIPO
4.1.4 P̲a̲c̲k̲a̲g̲e̲ ̲D̲a̲t̲a̲
Data areas used by LOG package:
a) Internal data area for temporary storage of log
records and status information.
b) On line and off line disk area used to store log
records.
4.1.5 C̲o̲m̲m̲o̲n̲ ̲D̲a̲t̲a̲
a) 1000 words data area common with TEP used as print
out buffer for up to 50 log records each 40 bytes
wide.
b) INIT-file used for package initialize.
4.1.5.1 L̲o̲g̲ ̲R̲e̲c̲o̲r̲d̲ ̲D̲a̲t̲a̲
Log record data includes a log record parameter list
and a view of the different log record formats and
types.
4.1.5.1.1 L̲o̲g̲ ̲R̲e̲c̲o̲r̲d̲ ̲P̲a̲r̲a̲m̲e̲t̲e̲r̲s̲
The list of log record parameters include a parameter
no. and the max. parameter size in memory. Numeric
only parameters (N) are in hexadecimal.
NO PARAMETER SIZE
IN
BYTE
̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲
̲ ̲ ̲ ̲
1 Log Type (main type + sub. type) N 1
2 Terminal/device designator 3
3 Transaction serial number N 2
4 Format ID N 2
5 Log time N 4
6 1 Item reference ID 2
7 Exit cause N 1
8 Classification 1
9 Special handling cat. 1
10 Start time of transaction N 4
11 2 Item reference ID 2
12 Month + day (general time) N 4
13 Decision code 1
14 User ID 3
15 Problem type 1
16 Document control number N 4
17 Channel designator 3
18 Channel serial number N 2
19 Message type 1
20 Precedence 1
21 Calling station 9
22 Station serial number N 2
NO PARAMETER SIZE
IN
BYTE
̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲
̲ ̲ ̲ ̲
23 Filing time N 4
24 SICs 9
25 Expected channel serial number N 2
26 Log record length N 1
27 Log record serial number N 2
28 Time stamp send N 4
4.1.5.1.2 L̲o̲g̲ ̲R̲e̲c̲o̲r̲d̲ ̲T̲y̲p̲e̲s̲
The log records are divided in main types and sub types.
Each type is assigned a type number.
The type numbers tell the receiver the origin of the
log records, and about the record format.
The main types are:
1) Log of user transactions
2) Log of security procedures
3) Log of supervisory transactions
4) Log of printer, PTP and OCR transactions
5) Log of reception/transmission of messages related
to NICS TARE, TRC, P-TO-P and PTR.
6) Log of CTS and CTS/A deletion.
The different log record types are shown on the following
pages.
The numbers in the main type record are the log record
parameter numbers.
The "X" in the sub type records indicates log data.
The empty boxes must be filled with 00 HEX.
Fig. 4.1.5.1.2-1 …01…Log of VDU User Transactions
Fig. 4.1.5.1.2-2…01…Log of VDU User Transactions
Fig. 4.1.5.1.2-3…01…Log of VDU User Transactions
Fig. 4.1.5.1.2-4…01…Log of VDU User Transactions
Fig. 4.1.5.1.2-5…01…Log of VDU User Transactions
Fig. 4.1.5.1.2-6…01…Log of VDU User Transactions
4.1.5.2 L̲o̲g̲ ̲T̲r̲a̲c̲e̲ ̲P̲a̲r̲a̲m̲e̲t̲e̲r̲s̲
The log trace function is implemented by LOG.
There are three types of log trace.
Type 1: Log time interval only.
Type 2: Channel trace + log time interval + sub-type
Subtype 1: Incoming messages
Subtype 2: Outgoing messages
Subtype 3: Channel discontinuity
Type 3: Terminal trace + log time interval + sub-type.
Subtype 1: Terminal procedure
Subtype 2: Message distribution control
Subtype 3: Message service
Subtype 4: Supervisor commands
Subtype 5: Security procedures
The log trace format (retrieval key format)
̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲
̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲
TYPE LOG TIME INTERVAL CHANNEL/TERMINAL DESIGNATOR
SUB TYPE
̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲
̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲
Byte: 1 8 3
1
4.1.6 I̲n̲t̲e̲r̲f̲a̲c̲e̲s̲
4.1.6.1 E̲x̲t̲e̲r̲n̲a̲l̲ ̲I̲n̲t̲e̲r̲f̲a̲c̲e̲s̲
There are no external interfaces to LOG.
4.1.6.2 P̲a̲c̲k̲a̲g̲e̲ ̲I̲n̲t̲e̲r̲f̲a̲c̲e̲s̲
An interface overview is given in fig. 4.1.6.2-1.
It covers all the LOG interfaces to other packages.
Each LOG interface is described in a separate chapter.
4.1.6.2.1 A̲p̲p̲l̲i̲c̲a̲t̲i̲o̲n̲s̲ ̲I̲n̲t̲e̲r̲f̲a̲c̲e̲s̲
Collection of log records is performed by means of
queue elements and the buffer function, implemented
by CSF. When a log record is generated, the application
must reserve a queue element and a buffer. The log
parameters are written in the reserved buffer and the
queue element is sent to the IN LOG queue with an answer
queue parameter.
When log records are received by LOG, the log time
and log record serial number is added to the log record
parameters.
An acknowledge queue element will be sent to the application
answer queue, when the log record is validated and
checkpointed.
Fig. 4.1.6.2-1…01…LOG Interface
Queue names used in fig. 4.1.6.2-1.
Q 1 IN LOG Incoming log queue
Q 2 APP ANSWQ Application answer queue
Q 3 IN COMM Incoming command queue
Q 4 COMM RESP Command response queue
Q 5 CHKP RESP Checkpoint response queue
Q 6 CTQ Checkpoint transmission
queue
Q 7 30 S TIM 30 seconds time out queue
Q 8 10MTIM 10 minutes time out queue
Q 9 RRQ Retrieval request queue
Q 10 RET RESP Retrieve response queue
Q 11 ISQ Incoming storage queue
Q 12 TRACE COMM Trace command queue
Q 13 SPQ Supervisor print queue
Q 14 SLPQ Supervisor log print queue
4.1.6.2.2 I̲n̲t̲e̲r̲f̲a̲c̲e̲ ̲t̲o̲ ̲S̲S̲C̲
The incoming log records are transferred to the checkpoint
transmission queue for checkpointing.
SSC returns a completion code to the checkpoint response
queue.
4.1.6.2.3 I̲n̲t̲e̲r̲f̲a̲c̲e̲ ̲t̲o̲ ̲T̲E̲P̲
The incoming log records are transferred to TEP for
printout. The log records are copied in a shared data
area and references to the log records in the data
area are queued in the supervisor log print queue.
If the number of unprinted log records exceed 50, then
additional incoming log records will be skipped. The
first entered queue element after one or more skipped
log records will be marked.
The stored log records may be traced on TEP request.
Trace requests and trace keys are transferred to LOG
by means of queue elements and the buffer function
implemented by CSF.
The traced log records will be copied in a CIF and
the CIF-ref. and a completion code is returned in a
queue element to the supervisor print queue.
If the requested log records cannot be traced then
a completion code is returned.
4.1.6.2.4 I̲n̲t̲e̲r̲f̲a̲c̲e̲ ̲t̲o̲ ̲C̲S̲F̲ ̲a̲n̲d̲ ̲M̲M̲S̲
The interface is implemented by means of monitor calls.
For further description please refer to CPS/ICD/009.
4.1.6.2.5 I̲n̲t̲e̲r̲f̲a̲c̲e̲ ̲t̲o̲ ̲S̲A̲R̲
The interface to SAR is implemented with queues. When
LOG requests SAR to catalogue and store a CIF, the
CIF-ref. and a SAR log type indicator is sent to the
SAR incoming storage queue.
A queue interface is also used when LOG request SAR
for a CIF-ref. A time interval and a SAR log type
indicator is sent to the SAR retrieval request queue.
A completion code or CIF-ref. is returned to the retrieve
response queue.
4.1.6.3 S̲u̲b̲-̲P̲a̲c̲k̲a̲g̲e̲ ̲I̲n̲t̲e̲r̲f̲a̲c̲e̲s̲
There is no interface between the two sub-packages.
4.2 S̲U̲B̲-̲P̲A̲C̲K̲A̲G̲E̲ ̲S̲P̲E̲C̲I̲F̲I̲C̲A̲T̲I̲O̲N̲S̲
The LOG package is divided in two independent sub-packages.
The sub-packages are : 4.2.1) Log collect sub-package
4.2.2) Log trace sub-package
The sub-packages are described in chapter 4.2.1 and
4.2.2.
4.2.1 L̲o̲g̲ ̲C̲o̲l̲l̲e̲c̲t̲ ̲S̲u̲b̲-̲P̲a̲c̲k̲a̲g̲e̲ ̲S̲p̲e̲c̲i̲f̲i̲c̲a̲t̲i̲o̲n̲
The log collect sub-package includes log record collection
and storage.
4.2.1.1 F̲u̲n̲c̲t̲i̲o̲n̲a̲l̲ ̲S̲p̲e̲c̲i̲f̲i̲c̲a̲t̲i̲o̲n̲
The functions for the log collect sub-package consists
of four main functions and a number of sub functions.
The main functions are:
1) Log record collection
2) Log record checkpointing
3) Log record printout
4) Log record store
4.2.1.1.1 L̲o̲g̲ ̲R̲e̲c̲o̲r̲d̲ ̲C̲o̲l̲l̲e̲c̲t̲i̲o̲n̲
Log records transferred to the incoming log queue are
validated. The validation includes syntax check and
format check. If an error is found, a completion code
is returned and the log record is deleted.
If a log passes the validation, it is stored in the
internal data area and the log record count is updated.
4.2.1.1.2 L̲o̲g̲ ̲R̲e̲c̲o̲r̲d̲ ̲C̲h̲e̲c̲k̲ ̲P̲o̲i̲n̲t̲i̲n̲g̲
A log record count (0-4) and a checkpoint serial number
(0-255) is added to the successful validated log records.
The two numbers are called log record serial number.
The log records are sent to SSC for checkpointing
and a completion code is awaited.
When the completion code is received from SSC, a completion
code will be returned to the log record sender.
4.2.1.1.3 L̲o̲g̲ ̲R̲e̲c̲o̲r̲d̲ ̲P̲r̲i̲n̲t̲o̲u̲t̲
The successfull validated log records are sent to the
common print-out data area and a pointer is sent to
the TEP supervisor log print queue.
Before a log record is sent for print out, the number
of unprinted log records is checked. If more than
49 log records are queued for print-out, then the function
will be skipped.
The first queued log record after the skip function
will be marked.
4.2.1.1.4 L̲o̲g̲ ̲R̲e̲c̲o̲r̲d̲ ̲S̲t̲o̲r̲e̲
The internally stored log records will be appended
to the present used CIF-ref. when one of two things
occur:
a) The number of stored log records reaches five.
b) A 30 sec. time out is received.
When the collected log records are appended, the log
record count is reset.
The present used CIF will be closed and catalogued
when a 10 min. time out occur. A new CIF will then
be created for online storage.
Fig. 4.2.1.1-1…01…Functional Breakdown
4.2.1.2 S̲o̲f̲t̲w̲a̲r̲e̲ ̲S̲t̲r̲u̲c̲t̲u̲r̲e̲
To describe the interrelationship between the used
software modules and define their level in the breakdown
hierarchy, an overview chart is shown. The box reference
numbers are used in the HIPO charts.
The timing between the software modules is achieved
by using coroutines, semaphores and flags.
The collect sub-package consists of three coroutines
implemented by the Coroutine Monitor.
Fig. 4.2.1.2-1…01…Software Structure
4.2.1.3 D̲a̲t̲a̲ ̲F̲l̲o̲w̲ ̲a̲n̲d̲ ̲C̲o̲n̲t̲r̲o̲l̲ ̲L̲o̲g̲i̲c̲
The data flow during collection and storage of log
records is presented in HIPO charts.
The numbers placed inside the boxes refer to a further
breakdown shown later in this document.
The coroutine interrelationship is shown in a block
diagram.
4.2.1.3.1 C̲o̲n̲t̲r̲o̲l̲ ̲L̲o̲g̲i̲c̲
4.2.1.3.1.1 C̲o̲r̲r̲o̲u̲t̲i̲n̲e̲ ̲I̲n̲t̲e̲r̲r̲e̲l̲a̲t̲i̲o̲n̲s̲h̲i̲p̲
The semaphores used between the three coroutines are
explained in the block diagram fig. 4.2.1.3.1-1.
S̲e̲m̲a̲p̲h̲o̲r̲e̲ ̲1̲
During the initialize procedure, executed by the collect
coroutine on SSC request, the SEM 1 counter is preset
to +5.
WAIT SEM 1 - Collect coroutine.
When a log record is received from
the incoming log queue, a SEM 1 waiting
point is reached.
SIGNAL SEM 1 - CHKP Response coroutine.
When a checkpoint response is received
from the checkpoint response queue
SEM 1 is signalled.
S̲e̲m̲a̲p̲h̲o̲r̲e̲ ̲2̲
WAIT SEM 2 - Store coroutine.
When a new CIF is created, store
coroutine waits on SEM 2 before storing
the old CIF.
SIGNAL SEM 2 - Collect coroutine.
After a CIF switch over, SEM 2 is
signalled.
F̲l̲a̲g̲ ̲1̲
SET FLAG 1 - Store coroutine.
Flag 1 indicates that a new CIF is
ready and that Store coroutine is
waiting.
RESET FLAG 1 - Collect coroutine.
The flag 1 is reset when a CIF switch-over
is executed.
F̲l̲a̲g̲ ̲2̲
SET FLAG 2 - Collect coroutine.
Flag 2 is set when the print-out queue
is full.
RESET FLAG 2 - Store coroutine.
When a log record is queued for print
out flag 2 is reset.
Fig. 4.2.1.3.2.1…01…Semaphore Control Logic …86…1 …02… …02… …02… …02…
4.2.1.3.2 D̲a̲t̲a̲ ̲F̲l̲o̲w̲
On the following pages a presentation of the Data Flow
is given on HIPO charts.
11 pages HIPOs
4.2.1.4 S̲u̲b̲-̲P̲a̲c̲k̲a̲g̲e̲ ̲D̲a̲t̲a̲
In this section, the data components for the Collect
sub-package is described.
a) Temporary storage of:
1) OLD CIF-ref. The OLD CIF-ref. is the presently
used CIF. Used for log record
storage.
2) NEW CIF-Ref. Is an empty CIF ready for use.
3) LOG RECORDS Is a log record collection
area. The area is also used
for storage of log time, record-count
and CONC-count.
4) BUFFER POINTER The buffer pointer contains
the key to a CSF buffer
5) CHECK POINT The checkpoint table contains
TABLE up to five queue references
used when completion codes
are returned to the application.
The CONC COUNT for the checkpointed
log records is also included
in the table. It is used to
identify the received completion
codes from SSC.
6) RECORD COUNT 0-4 The count is identical
to the number of internally
collected log records.
7) CONC COUNT 0-255 This count is used when
the checkpointed log records
are recovered.
8) FLAGS The flags used in COLLECT are
described in 4.2.1.3.1.1.
b) Permanent storage of log records and CIF-index
on disk.
c) Data used under record validation.
Validate parameters contain a description of the
different log record formats and parameter ranges.
d) Common Buffer Area:
This area is used to queue up to 50 log records
for print out to the TEP package.
4.2.1.5 S̲u̲b̲-̲P̲a̲c̲k̲a̲g̲e̲ ̲I̲n̲t̲e̲r̲f̲a̲c̲e̲
There is no interface between the two LOG sub-packages.
4.2.2 L̲o̲g̲ ̲T̲r̲a̲c̲e̲ ̲S̲u̲b̲-̲P̲a̲c̲k̲a̲g̲e̲ ̲S̲p̲e̲c̲i̲f̲i̲c̲a̲t̲i̲o̲n̲
4.2.2.1 F̲u̲n̲c̲t̲i̲o̲n̲a̲l̲ ̲S̲p̲e̲c̲i̲f̲i̲c̲a̲t̲i̲o̲n̲
The main functions for the log trace are:
1) Receive and validate trace keys.
2) Retrieve pertinent log CIFs
3) Trace log records
4) Return traced log records
The functional breakdown is shown in fig. 4.2.2.1-1
4.2.2.1.1 R̲e̲c̲e̲i̲v̲e̲ ̲a̲n̲d̲ ̲V̲a̲l̲i̲d̲a̲t̲e̲ ̲T̲r̲a̲c̲e̲ ̲K̲e̲y̲s̲
Retrieval keys transferred to the trace command queue
are validated. The validation includes syntax check
and format check. If any errors are found, the trace
action is terminated and a completion code returned
to the trace response queue. The log trace format
is shown in section 4.1.5.2.
4.2.2.1.2 R̲e̲t̲r̲i̲e̲v̲e̲ ̲P̲e̲r̲t̲i̲n̲e̲n̲t̲ ̲L̲o̲g̲ ̲C̲I̲F̲s̲
The pertinent log CIF-refs are retrieved from SAR.
Inputs are log time intervals computed by Log Trace
Sub-Package and SAR log type indicator. The log records
are read from the log CIF and stored in the log trace
data area.
4.2.2.1.3 T̲r̲a̲c̲e̲ ̲L̲o̲g̲ ̲R̲e̲c̲o̲r̲d̲s̲
The log records in the log trace data area are traced
one by one. The log record parameters are compared
with the trace keys. The wanted log records are stored
in the log trace CIF for later print-out.
4.2.2.1.4 R̲e̲t̲u̲r̲n̲ ̲T̲r̲a̲c̲e̲d̲ ̲L̲o̲g̲ ̲R̲e̲c̲o̲r̲d̲s̲
When log trace is completed, the log trace CIF-ref
and a completion code is returned to the TEP response
queue.
Fig. 4.2.2.1-1…01…Functional Breakdown
4.2.2.2 S̲o̲f̲t̲w̲a̲r̲e̲ ̲S̲t̲r̲u̲c̲t̲u̲r̲e̲
The Trace Sub-package is one of the coroutines in the
LOG package.
To describe the interrelationship between the used
software modules and to define their level in breakdown
hierarchy, an overview chart is shown in fig. 4.2.2.2-1.
Fig. 4.2.2.2-1…01…Software Structure
4.2.2.3 D̲a̲t̲a̲ ̲F̲l̲o̲w̲ ̲a̲n̲d̲ ̲C̲o̲n̲t̲r̲o̲l̲ ̲L̲o̲g̲i̲c̲
The data flow during log trace is presented in HIPO
charts.
4.2.2.3.1 D̲a̲t̲a̲ ̲F̲l̲o̲w̲
The numbers placed inside the boxes in the HIPO charts
refer to a further breakdown.
7 pages HIPOs
4.2.2.4 S̲u̲b̲-̲P̲a̲c̲k̲a̲g̲e̲ ̲D̲a̲t̲a̲
In this section, the data components for the Trace
sub-package is described.
a) T̲e̲m̲p̲o̲r̲a̲r̲y̲ ̲s̲t̲o̲r̲a̲g̲e̲ ̲o̲f̲:
1) BUFFER POINTER The Buffer pointer is a key
to the trace parameters.
2) CIF-STATUS Is a number indicating the
not traced log records in a
log CIF
3) LOG CIF-ref. Is a reference to the present
traced log CIF.
4) LOG COUNT The number of untraced log
records in a TRACE CIF PART.
5) LOG RECORDS A number of log records also
called TRACE CIF PART.
6) RECORD COLLECT Collection of traced and accepted
log records.
7) TIME INTERVAL Is generated from the time
interval given in the trace
keys.
8) TRACE CIF The traced and accepted log
records are stored in the TRACE
CIF.
9) TRACE COUNT The number of successfull traced
log records collected in main
memory.
10) TRACE CIF-ref. Is a reference to the trace
CIF.
11) TRACE COMPLETE A flag indicating that the
FLAG time interval limit is exceeded
12) TRACE KEYS The trace format is described
in section 4.1.5.2.
4.2.2.5 S̲u̲b̲-̲P̲a̲c̲k̲a̲g̲e̲ ̲I̲n̲t̲e̲r̲f̲a̲c̲e̲
There is no interface between the two LOG sub-packages.
…86…1 …02… …02… …02… …02… …02…
4.3 M̲E̲M̲O̲R̲Y̲ ̲L̲A̲Y̲O̲U̲T̲
Program Area: ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲
Collect 1000
Words
̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲
Trace 1000
Words
̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲
Data Area: ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲
Collect 300
Words
̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲
Trace 600
Words
̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲
Buffer Area: ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲
Buffer
1000
Words
Shared with TEP
̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲
Control Area: ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲
Coroutine 100
Words
Control
̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲
̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲
Process 1000
Words
Control
̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲ ̲
The max. used online mirroed disk area are:
- 128 K words for one busy hour
- 640 K words for 24 hours.
The max. used online disk area are:
- 24 M words for 30 days.
