System and method for logging and replaying asynchronous events
A system, method, and computer-accessible medium for logging and replaying asynchronous events are disclosed. One or more asynchronous events occurring during execution of a first instance of a computer program are logged. In logging the asynchronous events, a respective location in the execution of the first instance at which each of the one or more asynchronous events occurs is determined. A respective synchronous event preceding each asynchronous event is also determined. The asynchronous events are replayed during execution of a second instance of the computer program. In replaying each asynchronous event, the second instance is instrumented at the respective location during the execution of the second instance after detecting the preceding synchronous event.
1. A method comprising:
using a computer to perform:
logging one or more asynchronous events occurring during execution of a first instance of a computer program, comprising, for each of the one or more asynchronous events:
determining a respective location in the execution of the first instance at which the asynchronous event occurs; and
determining a respective synchronous event preceding the asynchronous event, wherein the respective synchronous event preceding the asynchronous event comprises a nearest synchronous event preceding the location at which the asynchronous event occurs in the execution of the first instance; and
replaying the one or more asynchronous events during execution of a second instance of the computer program, comprising, for each of the one or more asynchronous events:
detecting the respective synchronous event preceding the asynchronous event in the execution of the second instance; and
instrumenting the second instance at the respective location during the execution of the second instance after detecting the respective synchronous event, wherein instrumenting the second instance comprises:
determining an interval between the nearest synchronous event and the asynchronous event in the execution of the first instance;
if the interval is smaller than a threshold, inserting a breakpoint in the second instance; and
if the interval is not smaller than the threshold, entering a pre-breakpoint state to delay insertion of the breakpoint in the second instance.
2. The method of claim 1 ,
wherein logging the one or more asynchronous events further comprises, for each of the one or more asynchronous events, determining an original set of context values at the respective location in the execution of the first instance;
wherein replaying the one or more asynchronous events further comprises determining a current set of context values at each breakpoint during the execution of the second instance; and
wherein replaying the one or more asynchronous events further comprises, for each of the one or more asynchronous events, calling a signal handler function for the asynchronous event if the original set of context values for the asynchronous event match the current set of context values at the breakpoint for the asynchronous event.
3. The method of claim 1 ,
wherein replaying the one or more asynchronous events further comprises separating the second instance of the computer program into a plurality of execution blocks, wherein the plurality of execution blocks are separated by the synchronous events;
wherein instrumenting the second instance comprises, for each of the one or more asynchronous events, instrumenting the second instance upon entry to the execution block comprising the asynchronous event.
4. The method of claim 1 ,
wherein each of the respective locations in the execution of the first instance comprises a respective value of a performance counter at which the respective asynchronous event occurs.
5. The method of claim 1 ,
wherein the one or more asynchronous events comprise interactions between the first instance of the computer program and one or more external programs, wherein the interactions are initiated by the one or more external programs.
6. A non-transitory, computer-accessible storage medium comprising program instructions, wherein the program instructions are computer-executable to implement:
logging one or more asynchronous events occurring during execution of a first instance of a computer program, comprising, for each of the one or more asynchronous events:
determining a respective location in the execution of the first instance at which the asynchronous event occurs; and
determining a respective synchronous event preceding the asynchronous event, wherein the respective synchronous event preceding the asynchronous event comprises a nearest synchronous event preceding the location at which the asynchronous event occurs in the execution of the first instance; and
replaying the one or more asynchronous events during execution of a second instance of the computer program, comprising, for each of the one or more asynchronous events:
detecting the respective synchronous event preceding the asynchronous event in the execution of the second instance; and
instrumenting the second instance at the respective location during the execution of the second instance after detecting the respective synchronous event, wherein instrumenting the second instance comprises:
determining an interval between the nearest synchronous event and the asynchronous event in the execution of the first instance;
if the interval is smaller than a threshold, inserting a breakpoint in the second instance; and
if the interval is not smaller than the threshold, entering a pre-breakpoint state to delay insertion of the breakpoint in the second instance.
7. The non-transitory, computer-accessible storage medium of claim 6 ,
wherein logging the one or more asynchronous events further comprises, for each of the one or more asynchronous events, determining an original set of context values at the respective location in the execution of the first instance;
wherein replaying the one or more asynchronous events further comprises determining a current set of context values at each breakpoint during the execution of the second instance; and
wherein replaying the one or more asynchronous events further comprises, for each of the one or more asynchronous events, calling a signal handler function for the asynchronous event if the original set of context values for the asynchronous event match the current set of context values at the breakpoint for the asynchronous event.
8. The non-transitory, computer-accessible storage medium of claim 6 ,
wherein replaying the one or more asynchronous events further comprises separating the second instance of the computer program into a plurality of execution blocks, wherein the plurality of execution blocks are separated by the synchronous events;
wherein instrumenting the second instance comprises, for each of the one or more asynchronous events, instrumenting the second instance upon entry to the execution block comprising the asynchronous event.
9. The non-transitory, computer-accessible storage medium of claim 6 ,
wherein each of the respective locations in the execution of the first instance comprises a respective value of a performance counter at which the respective asynchronous event occurs.
10. A system comprising:
a first server comprising a first CPU and a first memory, wherein the first memory stores a first instance of a computer program which is executable by the first CPU, and wherein the first memory stores first program instructions which are executable by the first CPU to:
log one or more asynchronous events occurring during execution of the first instance of the computer program, comprising, for each of the one or more asynchronous events:
determining a respective location in the execution of the first instance at which the asynchronous event occurs; and
determining a respective synchronous event preceding the asynchronous event, wherein the respective synchronous event preceding the asynchronous event comprises a nearest synchronous event preceding the location at which the asynchronous event occurs in the execution of the first instance; and
a second server comprising a second CPU and a second memory, wherein the second memory stores a second instance of the computer program which is executable by the second CPU, and wherein the second memory stores second program instructions which are executable by the second CPU to:
replay the one or more asynchronous events during execution of the second instance of the computer program, comprising, for each of the one or more asynchronous events:
detecting the respective synchronous event preceding the asynchronous event in the execution of the second instance;
and
instrumenting the second instance at the respective location during the execution of the second instance after detecting the respective synchronous event, wherein instrumenting the second instance comprises:
determining an interval between the nearest synchronous event and the asynchronous event in the execution of the first instance;
if the interval is smaller than a threshold, inserting a breakpoint in the second instance; and
if the interval is not smaller than the threshold, entering a pre-breakpoint state to delay insertion of the breakpoint in the second instance.
11. The system of claim 10 ,
wherein, in logging the one or more asynchronous events, the first program instructions are further executable by the first CPU, for each of the one or more asynchronous events, to determine an original set of context values at the respective location in the execution of the first instance;
wherein, in replaying the one or more asynchronous events, the second program instructions are further executable by the second CPU to determine a current set of context values at each breakpoint during the execution of the second instance; and
wherein, in replaying the one or more asynchronous events, the second program instructions are further executable by the second CPU, for each of the one or more asynchronous events, to call a signal handler function for the asynchronous event if the original set of context values for the asynchronous event match the current set of context values at the breakpoint for the asynchronous event.
12. The system of claim 10 ,
wherein, in replaying the one or more asynchronous events, the second program instructions are further executable by the second CPU to separate the second instance of the computer program into a plurality of execution blocks, wherein the plurality of execution blocks are separated by the synchronous events;
wherein, in instrumenting the second instance, the second program instructions are further executable by the second CPU, for each of the one or more asynchronous events, to instrument the second instance upon entry to an execution block comprising the asynchronous event.
13. The system of claim 10 ,
wherein each of the respective locations in the execution of the first instance comprises a respective value of a performance counter at which the respective asynchronous event occurs.