JBPM-7266 Configure jBPM unit tests to use connection pooling #1217
Conversation
this is actually misconfiguration I'd say - since there is jta-data-source given in persistence.xml username and password in properties should anyway be ignored so I'd say let's remove it and rely only on data source instead. |
|
+1, will remove once we have something working |
|
@rsynek how are we doing with this one? do we wait for JWS team at the moment? |
|
@mswiderski just FYI @rsynek is on vacation till the ond of this week. |
|
@mswiderski yes, here is the issue I opened for this purpose https://issues.jboss.org/browse/JWS-1028 |
|
jenkins retest this |
|
jenkins retest this |
|
@rsynek can you try my commit to fix the interposed synchonization |
rsynek
force-pushed
the
baqe-439
branch
2 times, most recently
from
4305bb8
to
8e50eb1
Compare
|
Can one of the admins verify this patch? |
|
jenkins retest this |
| xaDataSource.getClass().getMethod("setPassword", new Class[]{String.class}).invoke(xaDataSource, password); | ||
| xaDataSource.getClass().getMethod("setUser", new Class[]{String.class}).invoke(xaDataSource, username); | ||
| } | ||
|
|
||
| if (!(className.startsWith("com.ibm.db2") || className.startsWith("com.sybase"))) { |
There was a problem hiding this comment.
Shouldn't this be rather else if then?
There was a problem hiding this comment.
There is already else branch on line https://github.com/rsynek/jbpm/blob/b5feebfc6fce286b5e66ef07b6fb80963bf0ec0b/jbpm-test-util/src/main/java/org/jbpm/test/util/PoolingDataSource.java#L156 being currently executed for H2 too. I will think about making this part more readable - right now I wanted to see how many failures this fix eliminates.
There was a problem hiding this comment.
A bit of refactoring we talked about - in the target (not yet used) component web-servers/narayana-tomcat#1
|
Can one of the admins verify this PR? Comment with 'ok to test' to start the build. |
|
jenkins retest this |
|
Failed probably due to jenkins being restarted. |
|
jenkins retest this |
|
The single failure is probably a random one; locally the test works. |
|
@rsynek so what’s the next step for it? |
|
@mswiderski the next step is to change the tomcat-dbcp 9.0-SNAPSHOT to stable version, once it's ready. I will also request a build of tomcat-jta with [1] and remove the boiler plate code from jbpm-test-util. This second step is not necessary, but nice to have. |
|
jenkins retest this |
…oOpTransactionManager.java
|
jenkins retest this (testing with narayana-jta-5.8.1) |
|
Closing to merge https://github.com/kiegroup/drools/pull/1939 first. A new PR will replace this one soon. |
Provides integration between jBPM engine, Narayana and Tomcat DBCP. Changes introduced by this PR have been already tested with various DBs and JEE containers.
Depends on https://github.com/kiegroup/drools/pull/1939
[DO NOT MERGE] - waiting for tomcat-dbcp stable release