Computer resource leak detection
Various embodiments discussed herein enable intelligent resource leak detection and associated user interface. In some aspects, if a kernel sends a tracing event (e.g., an ETW event) after a process or application has been requested to be shut down but before the process has actually been shut down, a resource leak can be detected. This is because the application should have requested the reference indicator to be removed in memory before the tracing event, as opposed to the kernel acting alone without an application request to remove the reference indicator. If a reference indicator has not been requested to be removed by the time this tracing event occurs, then it is indicative of the application never having sent a request to remove the reference indicator, which is indicative of a resource leak.
1. A system comprising: a processor; and a computer storage media storing computer-useable instructions that, when used by the processor, causes the processor to perform a method, the method comprising: detecting a process launch; determining a plurality of reference indicators that are generated and terminated for the process while the process is running, each of the plurality of reference indicators referencing a respective computing resource; generating, for display on a user interface, first values indicating the plurality of reference indicators that are generated and terminated; determining that a set of reference indicators that are generated but are not terminated exceed a performance issue threshold associated with a resource leak; and based at least in part on the determining that the set of reference indicators exceed the performance issue threshold, generating, for display on the user interface, a second value indicating a performance issue associated with the performance issue threshold.
2. The system of claim 1 , wherein the method further comprises: receiving a first indication that the process has been requested to be shut down; based on the first indication, receiving a second indication that an operating system kernel will remove the set of reference indicators; and detecting a resource leak based at least in part on the first indication and the second indication.
3. The system of claim 2 , wherein the performance issue threshold is a count threshold such that a quantity of the set of reference indicators that exceeds the count threshold is indicative of too many reference indicators that are generated but not removed.
4. The system of claim 3 , the method further comprising generating a message indicating the set of reference indicators have exceeded the count threshold and is associated with the resource leak.
5. The system of claim 1 , wherein the reference indicator is one of a graphics object identifier and a user handle.
6. The system of claim 1 , the method further comprising generating a mapping for a reference indicator to a category that the reference indicator belongs to and providing, for display to the user interface, the category as a column identifier.
7. The system of claim 1 , wherein the determining that the set of reference indicators that are generated but not terminated occurs while the process is still running.
8. The system of claim 1 , the method further comprising determining a location in a call stack for which the performance issue is detected and cause display to the user interface of the location.
9. The system of claim 1 , wherein the determining that the set of reference indicators is generated but not terminated is only an indication that there might be the resource leak, and wherein the resource leak is not definitively determined until the process is shutdown.