Hang detection and remediation in a multi-threaded application process

A computer-implemented method for detecting non-callable external component application programming interfaces, the computer-implemented method comprising:

• determining, by a monitoring thread of a computer, whether a first function call stack of a worker thread in a multi-threaded application of the computer matches a second function call stack of the worker thread;
• responsive to the monitoring thread determining that the first function call stack matches the second function call stack of the worker thread, identifying, by the monitoring thread, an external component application programming interface (API) corresponding to the worker thread from a function call stack of the worker thread;
• marking, by the monitoring thread, the external component API corresponding to the worker thread as non-callable in an API state map;
• marking, by the monitoring thread, the worker thread as being in a hang state;
• terminating, by the monitoring thread, the worker thread in the hang state as a remediation action step to maintain performance of the computer;
• reading, by the monitoring thread, a configuration properties file of the multi-threaded application to identify a specified set of external component APIs to monitor and maximum execution time limits corresponding to the specified set of external component APIs;
• generating, by the monitoring thread, a worker thread state map that records execution start times of worker threads and corresponding external component APIs with maximum execution time limits;
• and
• performing, by the monitoring thread, an analysis of the worker thread state map at predefined time intervals.

A computer-implemented method for detecting non-callable external component application programming interfaces, the computer-implemented method comprising:

• determining, by a monitoring thread of a computer, whether a first function call stack of a worker thread in a multi-threaded application of the computer matches a second function call stack of the worker thread;
• responsive to the monitoring thread determining that the first function call stack matches the second function call stack of the worker thread, identifying, by the monitoring thread, an external component application programming interface (API) corresponding to the worker thread from a function call stack of the worker thread;
• marking, by the monitoring thread, the external component API corresponding to the worker thread as non-callable in an API state map;
• marking, by the monitoring thread, the worker thread as being in a hang state;
• terminating, by the monitoring thread, the worker thread in the hang state as a remediation action step to maintain performance of the computer;
• determining, by the worker thread, that the external component API corresponding to a client work request is specified to be monitored in a configuration properties file of the multi-threaded application;
• updating, by the worker thread, a worker thread state map with an entry that includes an identifier, an execution start time, and a maximum execution time limit for the external component API;
• invoking, by the worker thread, the external component API to execute the client work request;
• determining, by the worker thread, whether execution of the external component API exceeded the maximum execution time limit based on the execution start time;
• and
• responsive to the worker thread determining that the execution of the external component API did not exceed the maximum execution time limit based on the execution start time, removing, by the worker thread, the entry that includes the identifier, the execution start time, and the maximum execution time limit for the external component API from the worker thread state map.

A computer system for detecting non-callable external component application programming interfaces, the computer system comprising:

• a bus system;
• a storage device connected to the bus system, wherein the storage device stores program instructions;
• and
• a processor connected to the bus system, wherein the processor executes the program instructions to:
• determine, by a monitoring thread of the computer system, whether a first function call stack of a worker thread in a multi-threaded application of the computer system matches a second function call stack of the worker thread;
• identify, by the monitoring thread, an external component application programming interface (API) corresponding to the worker thread from a function call stack of the worker thread in response to the monitoring thread determining that the first function call stack matches the second function call stack of the worker thread;
• mark, by the monitoring thread, the external component API corresponding to the worker thread as non-callable in an API state map;
• mark, by the monitoring thread, the worker thread as being in a hang state;
• and
• terminate, by the monitoring thread, the worker thread in the hang state as a remediation action step to maintain performance of the computer system,
• wherein the processor further executes the program instructions to:
• determine, by the monitoring thread, whether the worker thread exceeded a transaction time out limit of a client application;
• determine, by the monitoring thread, whether a worker thread state map indicates the external component API corresponding to the worker thread in response to the monitoring thread determining that the worker thread did exceed the transaction time out limit of the client application;
• take, by the monitoring thread, the first function call stack of the worker thread at a first time and take, by the monitoring thread, the second function call stack of the worker thread at a second time in response to the monitoring thread determining that the worker thread state map is silent regarding the external component API corresponding to the worker thread;
• and
• perform, by the monitoring thread, a comparison of text of the first function call stack to text of the second function call stack of the worker thread.

A computer system for detecting non-callable external component application programming interfaces, the computer system comprising:

• a bus system;
• a storage device connected to the bus system, wherein the storage device stores program instructions;
• and
• a processor connected to the bus system, wherein the processor executes the program instructions to:
• determine, by a monitoring thread of the computer system, whether a first function call stack of a worker thread in a multi-threaded application of the computer system matches a second function call stack of the worker thread;
• identify, by the monitoring thread, an external component application programming interface (API) corresponding to the worker thread from a function call stack of the worker thread in response to the monitoring thread determining that the first function call stack matches the second function call stack of the worker thread;
• mark, by the monitoring thread, the external component API corresponding to the worker thread as non-callable in an API state map;
• mark, by the monitoring thread, the worker thread as being in a hang state;
• and
• terminate, by the monitoring thread, the worker thread in the hang state as a remediation action step to maintain performance of the computer system,
• wherein the processor further executes the program instructions to:
• read, by the monitoring thread, a configuration properties file of the multi-threaded application to identify a specified set of external component APIs to monitor and maximum execution time limits corresponding to the specified set of external component APIs;
• generate, by the monitoring thread, a worker thread state map that records execution start times of worker threads and corresponding external component APIs with maximum execution time limits;
• and
• perform, by the monitoring thread, an analysis of the worker thread state map at predefined time intervals.

A computer program product for detecting non-callable external component application programming interfaces, the computer program product comprising a computer-readable storage medium having program instructions embodied therewith, the program instructions executable by a computer to cause the computer to:

• determining, by a monitoring thread of the computer, whether a first function call stack of a worker thread in a multi-threaded application of the computer matches a second function call stack of the worker thread;
• responsive to the monitoring thread determining that the first function call stack matches the second function call stack of the worker thread, identifying, by the monitoring thread, an external component application programming interface (API) corresponding to the worker thread from a function call stack of the worker thread;
• marking, by the monitoring thread, the external component API corresponding to the worker thread as non-callable in an API state map;
• marking, by the monitoring thread, the worker thread as being in a hang state;
• terminating, by the monitoring thread, the worker thread in the hang state as a remediation action step to maintain performance of the computer;
• determining, by the monitoring thread, whether the worker thread exceeded a transaction time out limit of a client application;
• responsive to the monitoring thread determining that the worker thread did exceed the transaction time out limit of the client application, determining, by the monitoring thread, whether a worker thread state map indicates the external component API corresponding to the worker thread;
• responsive to the monitoring thread determining that the worker thread state map is silent regarding the external component API corresponding to the worker thread, taking, by the monitoring thread, the first function call stack of the worker thread at a first time and taking, by the monitoring thread, the second function call stack of the worker thread at a second time;
• and
• performing, by the monitoring thread, a comparison of text of the first function call stack to text of the second function call stack of the worker thread.

A computer program product for detecting non-callable external component application programming interfaces, the computer program product comprising a computer-readable storage medium having program instructions embodied therewith, the program instructions executable by a computer to cause the computer to:

• determining, by a monitoring thread of the computer, whether a first function call stack of a worker thread in a multi-threaded application of the computer matches a second function call stack of the worker thread;
• responsive to the monitoring thread determining that the first function call stack matches the second function call stack of the worker thread, identifying, by the monitoring thread, an external component application programming interface (API) corresponding to the worker thread from a function call stack of the worker thread;
• marking, by the monitoring thread, the external component API corresponding to the worker thread as non-callable in an API state map;
• marking, by the monitoring thread, the worker thread as being in a hang state;
• terminating, by the monitoring thread, the worker thread in the hang state as a remediation action step to maintain performance of the computer;
• reading, by the monitoring thread, a configuration properties file of the multi-threaded application to identify a specified set of external component APIs to monitor and maximum execution time limits corresponding to the specified set of external component APIs;
• generating, by the monitoring thread, a worker thread state map that records execution start times of worker threads and corresponding external component APIs with maximum execution time limits;
• and
• performing, by the monitoring thread, an analysis of the worker thread state map at predefined time intervals.