[ODLPARENT-179] SFT seems to hang for ~30 seconds after finishing the test Created: 13/Dec/18  Updated: 19/Jun/20  Resolved: 19/Jun/20

Status: Resolved
Project: odlparent
Component/s: SFT
Affects Version/s: 4.0.2
Fix Version/s: 5.0.8, 6.0.8, 7.0.4

Type: Bug Priority: High
Reporter: Robert Varga Assignee: Robert Varga
Resolution: Done Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified


 Description   

The following output is quite common, for example in https://logs.opendaylight.org/releng/vex-yul-odl-jenkins-1/bgpcep-maven-verify-neon-mvn35-openjdk8/831/console-timestamp.log.gz :

16:26:04  [INFO] --- maven-surefire-plugin:2.22.0:test (default) @ odl-bgpcep-pcep ---
16:26:04  [INFO] Surefire report directory: /w/workspace/bgpcep-maven-verify-neon-mvn35-openjdk8/features/pcep/odl-bgpcep-pcep/target/surefire-reports
16:26:04  [INFO] 
16:26:04  [INFO] -------------------------------------------------------
16:26:04  [INFO]  T E S T S
16:26:04  [INFO] -------------------------------------------------------
16:26:04  [INFO] Running org.opendaylight.odlparent.featuretest.SingleFeatureTest
16:26:46  [INFO] Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 42.302 s - in org.opendaylight.odlparent.featuretest.SingleFeatureTest
16:27:17  [INFO] 
16:27:17  [INFO] Results:
16:27:17  [INFO] 
16:27:17  [INFO] Tests run: 1, Failures: 0, Errors: 0, Skipped: 0
16:27:17  [INFO] 
16:27:17  [INFO] 

Note that while the test ran for 42 seconds, an additional ~31 seconds is spent waiting for ... something ... to happen. The delay seems to fluctuate between 29-31 seconds, oftentimes it's not there at all.

This is extremely inefficient and hurts especially projects with a large number of features - as these delays pile up rather quickly.



 Comments   
Comment by Robert Varga [ 13/Dec/18 ]

This seems to be related to https://maven.apache.org/surefire/maven-surefire-plugin/examples/shutdown.html and JVM shutdown, which is capped at 30 seconds. Perhaps there is a non-daemon thread or a shutdown hook which is being processed.

https://issues.apache.org/jira/browse/SUREFIRE-1574 seems to be very much related and is targeted for surefire-maven-plugin-3.0.0.M4. pax-exam issue is at https://ops4j1.jira.com/projects/PAXEXAM/issues/PAXEXAM-920 .

Comment by Robert Varga [ 23/Mar/20 ]

pax-exam-4.13.3 contains the fix for PAXEXAM-920

Comment by Robert Varga [ 12/Jun/20 ]

This did not help, but surefire-3.0.0-M5 is just about to land, hence target this for $NEXT releases so we resolve it as soon as surefire releases.

Generated at Wed Feb 07 20:27:51 UTC 2024 using Jira 8.20.10#820010-sha1:ace47f9899e9ee25d7157d59aa17ab06aee30d3d.