In a microservices-based application, tracking errors by mapping traces to error stacks

1. A method of tracking errors in a system comprising microservices, the method comprising:

consolidating a plurality of ingested spans associated with a first duration of time into a plurality of traces, wherein each of the plurality of traces comprise a subset of the plurality of ingested spans;

for each trace of the plurality of traces:

mapping a respective trace to one or more error stacks computed for the respective trace and to one or more attributes determined for the respective trace; and

pairing each error stack computed from the respective trace with a set of associated attributes.

2. The method of claim 1 , further comprising:

reducing duplicate pairs of error stacks and associated attributes and maintaining a count value for each pair of error stack and associated attributes.

3. The method of claim 1 , further comprising:

reducing duplicate pairs of error stacks and associated attributes and maintaining a count value for each pair of error stack and associated attributes;

storing each pair of error stack and associated attributes with the count value in a database, wherein the count value is associated with a number of times each pair occurred during the first duration of time; and

storing each pair of error stack and associated attributes with a summary metric in the database, and wherein the database is operable to be queried by a client.

4. The method of claim 1 , further comprising:

repeating the consolidating, the mapping, and the pairing for each subsequent duration of time.

5. The method of claim 1 , wherein the one or more attributes determined for the respective trace pertain to a user request.

6. The method of claim 1 , wherein the one or more error stacks computed for the respective trace comprises an ordered list of span error frames, wherein a top-most frame in the ordered list is associated with a root cause error span, wherein each subsequent span error frame is associated with a parent span of a prior span in the ordered list, wherein a root cause error span comprises an error span associated with a microservice that originated an error, and wherein each span error frame in the ordered list comprises summary information associated with a respective span.

7. The method of claim 1 , further comprising:

reducing duplicate pairs of error stacks and associated attributes and maintaining a count value for each pair of error stack and associated attributes;

storing each pair of error stack and associated attributes with the count value in a database, wherein the count value is associated with a number of times each pair occurred during the first duration of time; and

providing a user a graphical user interface (GUI) to query the database.

8. The method of claim 1 , further comprising:

reducing duplicate pairs of error stacks and associated attributes and maintaining a count value for each pair of error stack and associated attributes;

storing each pair of error stack and associated attributes with the count value in a database, wherein the count value is associated with a number of times each pair occurred during the first duration of time; and

providing a graphical user interface (GUI) to query the database, wherein the GUI provides a user options to filter error information for each stored error stack across the associated attributes.

9. The method of claim 1 , further comprising:

prior to the consolidating, flagging spans, from the plurality of ingested spans, that are associated with probe data;

reducing duplicate pairs of error stacks and associated attributes and maintaining a count value for each pair of error stack and associated attributes;

storing each pair of error stack and associated attributes with the count value in a database, wherein the count value is associated with a number of times each pair occurred during the first duration of time;

rendering a graphical user interface (GUI) operable to submit a query to the database; and responsive to a user selection of an exclude option in the GUI, excluding the probe data from a result set displayed in response to the query.

10. The method of claim 1 , further comprising:

prior to the consolidating, excluding spans from the plurality of ingested spans that are associated with probe data.

11. The method of claim 1 , further comprising:

prior to the consolidating, excluding spans from the plurality of ingested spans that are associated with probe data, wherein the probe data is defined based on user-specified criteria entered using a GUI.

12. The method of claim 1 , further comprising:

prior to the consolidating, excluding spans from the plurality of ingested spans that are associated with probe data, wherein the probe data is defined based on user-specified criteria.

13. A non-transitory computer-readable medium having computer-readable program code embodied therein for causing a computer system to perform a method for tracking errors in a system comprising microservices, the method comprising:

consolidating a plurality of ingested spans associated with a first duration of time into a plurality of traces, wherein each of the plurality of traces comprise a subset of the plurality of ingested spans;

for each trace of the plurality of traces:

mapping a respective trace to one or more error stacks computed for the respective trace and to one or more attributes determined for the respective trace; and

pairing each error stack computed from the respective trace with a set of associated attributes.

14. The non-transitory computer-readable medium of claim 13 , wherein the method further comprises:

reducing duplicate pairs of error stacks and associated attributes and maintaining a count value for each pair of error stack and associated attributes.

15. The non-transitory computer-readable medium of claim 13 , wherein the method further comprises:

reducing duplicate pairs of error stacks and associated attributes and maintaining a count value for each pair of error stack and associated attributes;

storing each pair of error stack and associated attributes with the count value in a database, wherein the count value is associated with a number of times each pair occurred during the first duration of time; and

storing each pair of error stack and associated attributes with a summary metric in the database, and wherein the database is operable to be queried by a client.

16. The non-transitory computer-readable medium of claim 13 , wherein the method further comprises:

repeating the consolidating, the mapping, and the pairing for each subsequent duration of time.

17. The non-transitory computer-readable medium of claim 13 , wherein the method further comprises, wherein the one or more attributes determined for the respective trace pertain to a user request.

18. The non-transitory computer-readable medium of claim 13 , wherein the one or more error stacks computed for the respective trace comprises an ordered list of span error frames, wherein a top-most frame in the ordered list is associated with a root cause error span, wherein each subsequent span error frame is associated with a parent span of a prior span in the ordered list, wherein a root cause error span comprises an error span associated with a microservice that originated an error, and wherein each span error frame in the ordered list comprises summary information associated with a respective span.

19. A system for tracking errors in a network comprising microservices, the system comprising:

a processing device communicatively coupled with a memory and configured to:

consolidate a plurality of ingested spans associated with a first duration of time into a plurality of traces, wherein each of the plurality of traces comprise a subset of the plurality of ingested spans;

for each trace of the plurality of traces:

map a respective trace to one or more error stacks computed for the respective trace and to one or more attributes determined for the respective trace; and

pair each error stack computed from the respective trace with a set of associated attributes.

20. The system of claim 19 , wherein the processing device is further configured to:

reduce duplicate pairs of error stacks and associated attributes and maintaining a count value for each pair of error stack and associated attributes.