Closed askhatri closed 1 year ago
Merging #426 (9a1f31b) into master (8b2e29f) will decrease coverage by
36.78%
. The diff coverage is48.57%
.
@@ Coverage Diff @@
## master #426 +/- ##
=============================================
- Coverage 65.48% 28.71% -36.78%
+ Complexity 950 382 -568
=============================================
Files 103 103
Lines 6062 6084 +22
Branches 916 922 +6
=============================================
- Hits 3970 1747 -2223
- Misses 1541 3980 +2439
+ Partials 551 357 -194
Files | Coverage Δ | |
---|---|---|
...va/org/apache/livy/client/common/HttpMessages.java | 85.71% <100.00%> (-10.65%) |
:arrow_down: |
.../main/scala/org/apache/livy/sessions/Session.scala | 63.63% <100.00%> (-11.16%) |
:arrow_down: |
...e/livy/server/interactive/InteractiveSession.scala | 70.32% <85.71%> (+8.10%) |
:arrow_up: |
.../server/interactive/CreateInteractiveRequest.scala | 61.76% <33.33%> (-13.24%) |
:arrow_down: |
...server/interactive/InteractiveSessionServlet.scala | 55.69% <66.66%> (-3.66%) |
:arrow_down: |
...cala/org/apache/livy/sessions/SessionManager.scala | 58.92% <7.14%> (-23.43%) |
:arrow_down: |
... and 80 files with indirect coverage changes
:mega: We’re building smart automated test selection to slash your CI/CD build times. Learn more
LGTM, thanks for your contribution!
What changes were proposed in this pull request?
Currently, a Livy interactive session has a field called ttl, which kills the session if it has been idle for a given amount of time. However, here is the expected behaviour:
ttl: kills the session if it has been active for a certain duration, irrespective of idleness. idleTimeout: kills the session if it has been idle for the given duration. (The current TTL behaves in this manner.)
JIRA: https://issues.apache.org/jira/browse/LIVY-989
How was this patch tested?
Tested manually by creating interactive session with idle time and TTL.