Error recovery for multi-stage simultaneously running tasks
View Patent ↗A method of managing a server is described where a service performed by a task executing on the server is configured into a plurality of sub-tasks, which may be further configured into a plurality of subsidiary tasks. When a subsidiary task executes, the completion thereof is determined to be either a successful or an unsuccessful completion. The completion status of the ensemble of subsidiary tasks is assessed on each completion of a subsidiary task, and the failure of any subsidiary task to successfully complete is deemed to be a failure to complete of the server task. The failure to complete is reported to the requesting application without waiting for the completion of any pending subsidiary task.
1. A method of managing a server with a stateless connection to a user, the method comprising:
providing a server computer having a communications interface with a user computer; and
a memory system,
wherein responding to a request for a service from the user computer comprises:
receiving a user request from the user computer over a communications interface;
executing a task by the server computer to respond to the user request, the task comprising a supervisory task and a number of sub-tasks, each sub-task of the number of sub-tasks having one or more subsidiary tasks; and
configuring each sub-task to report a completion status of each subsidiary task and completion of the sub-task to the supervisory task each time a subsidiary task completes;
wherein a completion status of a subsidiary task is one of success or error, and the supervisory task determines the completion status of each subsidiary task at the completion time of the subsidiary task prior to completion of the sub-task and,
the supervisory task is configured to report an error in completion of the subsidiary task of the sub-task to the user computer if the completion status of the subsidiary task is error, wherein the error terminates or aborts execution and completion of the sub-task and other sequential subsidiary tasks of the sub-task;
wait for another completion status change when a number of completed sub-tasks is less than the number of sub-tasks; and
respond to the user request over the communications interface when a number of successfully completed sub-tasks is equal to the number of sub-tasks.
2. A computer program product, stored on a machine-readable non-transitory medium, comprising:
instructions executable by a computer to determine a status of a task program,
wherein the task program is configured as a plurality of sub-tasks and the sub-tasks are configured as a number of subsidiary tasks, which execute contemporaneously;
a supervisory sub-task of the task program monitors a completion status of each subsidiary task and the completion status is one of successful completion or error; and
wherein completion of each subsidiary task of the plurality of subsidiary tasks results in an evaluation of the completion status of the completed subsidiary task by the supervisory sub-task; and, the supervisory sub-task is configured to:
report the completion status of the task program as a failure when the completion status of error is reported by the subsidiary task, wherein the failure terminates or aborts execution of the supervisory task and any other subsidiary tasks of the supervisory task;
complete the task program when a number of subsidiary reporting successful completion is equal to the number of subsidiary tasks; and
wait for another completion status change when a number of subsidiary tasks reporting successful completion is less than the number of subsidiary tasks.
3. The method of claim 1 , wherein the subsidiary tasks are further configured to:
set completion flags in a status matrix when the subsidiary tasks are successfully completed; and
set error flags in the status matrix when an error is detected in the subsidiary tasks.
4. The method of claim 3 , wherein any set error flags in the status matrix terminates or aborts uncompleted sub-tasks.
5. The method of claim 3 , wherein the supervisory task is further configured to:
check the status matrix after completion of each of the sub tasks subsidiary tasks; and
terminate the task when any of the error flags are set in the status matrix.