International Business Machines Corporation

Stephen John Burghard

Andrew Wright

David J Harman

Neil William Leedham

Mark Andrew Woolley

Ian J Mitchell

Samuel Jonathan Smith

Philip R Lee

Mark L Todd

Brenda Hawkins

Daniel J Rivett

Arrangements described herein relate to performing diagnostic tracing of an executing application. A trace entry in trace data can be identified, the trace entry comprising a pointer that refers to a memory address. Whether a value that is, or has been, stored at the memory address is an erroneous value can be determined. Responsive to determining that the value that is, or has been, stored at the memory address is an erroneous value, the pointer can be indicated as being a suspicious value.

Performing diagnostic tracing of an executing application to identify suspicious pointer values

Creating additional trace entries by dynamically processing recently captured output data, working data, and input data to diagnose a software error. Integrating additional trace entries in chronological order with conventional trace entries into a single trace dataset for analysis.

Creating trace data from recent software output and activity

A system and computer program product for developing software in which the software comprises a plurality of programs. A change to a program is received. A data structure checking procedure may then be invoked. The changed program is parsed for a reference to a data structure. Other instances of the data structure are located in other programs within the software. The referenced data structure is compared to the located other instances of the data structure. A predefined action (such as notifying a programmer or correcting the inconsistencies) is performed in response to any detected differences between the referenced data structure and the located other instances of the data structure. These steps are repeated for all data structures within the changed program.

Maintaining consistency amongst data structures that are referenced within different programs

A method for developing software in which the software comprises a plurality of programs. A change to a program is received. A data structure checking procedure may then be invoked. The changed program is parsed for a reference to a data structure. Other instances of the data structure are located in other programs within the software. The referenced data structure is compared to the located other instances of the data structure. A predefined action (such as notifying a programmer or correcting the inconsistencies) is performed in response to any detected differences between the referenced data structure and the located other instances of the data structure. These steps are repeated for all data structures within the changed program.

A method of operating a transaction server for use with trace components, providing an execution cost for each trace component, receiving input selecting one or more trace components, running the transaction server with each selected trace component active, monitoring the execution cost of each selected trace component, and providing an updated execution cost for each trace component. The method can also be extended to provide an execution cost for one or more combinations of trace components monitoring the execution cost of any selected trace component combination, and also to provide an updated execution cost for each trace component combination. The method can further comprise maintaining a record of the execution cost of each selected trace component and transmitting the record to a different transaction server, thereby sharing knowledge between connected systems.

Trace monitoring

Data fields within a trace data set are interpreted using data field declarations of the data fields that each specify a data type definition of a respective data field. A data value of an interpreted data field is compared with the data type definition specified within a respective data field declaration for the interpreted data field within the trace date set. Based upon the comparison, a determination is made that the respective interpreted data field contains a data value that violates the data type definition specified within the respective data field declaration.

Trace value correlation with data field declarations

A method, apparatus and computer program for recording the core data of a computer process, the computer process comprising trace points and core data is provided for each such trace point. A first set of core data comprising an image of a memory for the computer process is stored in response to a first set of trace data being produced for the computer process for a first trace point. A second set of core data is stored in response to a second set of trace data being produced for the computer process for a second trace, where the second set of core data comprises a record of any change in current memory contents for the computer process with respect to the first set of core data.

Recording the core data of a computer process which provides trace data

A method for identifying a storage location for a requested storage address. The method includes receiving a request to view data at a storage address and determining the requested storage address corresponding to a plurality of storage locations. The method includes determining whether the requested storage address identifies memory related to a dump file being analyzed by a dump formatter. Then, in response to determining the requested storage address identifies memory related to the dump file being analyzed by the dump formatter, the method includes identifying one of the plurality of storage locations. The method includes directing the request to the identified storage location.

Identifying a storage location for a storage address requested during debugging

A transaction processing system comprising a transaction log, a log management policy, a profile and a log manager, and method for managing the transaction log are provided. The method comprises maintaining a transaction log of recoverable changes made by transaction processing tasks, each task having an associated transaction type; storing a log management policy including at least one log threshold and a plurality of possible actions associated therewith; and storing a profile of log resource usage characteristics of tasks for different types of transactions. Usage of the log by transaction processing tasks is monitored to determine when a log threshold is reached. The profile indicates the normal logging behavior of particular types of transaction by identifying ranges of acceptable values for a plurality of log resource usage characteristics. Such characteristics may include time taken for a task to complete, CPU resource consumed, number of log records made, etc. From a comparison of the profile behavior with the current log resource usage of a task which has been identified as holding up trimming of the log, likelihood that the identified task having entered an invalid loop and represents a 'bad' program, rather than simply validly taking a long time to complete can be inferred. Thus the most appropriate action can be selected and taken.

Transaction log management

Data fields within a trace data set are interpreted by a processor using a data field declaration of each data field that identifies a data type of the respective data field. An actual data value of the interpreted data field is compared with the data field declaration for each interpreted data field. Based upon each comparison, a determination is made as to whether the actual data value of the interpreted data field contains an erroneous data value based upon the data field declaration.

A system for tracking register and/or storage area assignments and warning a programmer of potential assignment errors. More specifically, the system tracks programmer assignments to registers and storage areas and determines if a register or storage area has been modified prior to a call to a program or process external to the source code being examined. The system notes whether a modified register or storage area is restored to its original value subsequent to the external call. If the register or storage area has not been restored, the system displays a warning of a possible assignment error.

Growing community of inventors

Portsmouth, United Kingdom

Stephen John Burghard