[
https://issues.apache.org/jira/browse/DERBY-4270?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Dag H. Wanvik updated DERBY-4270:
---------------------------------
Attachment: derby-4270.diff
Uploading a fix which adds an overload of JUnit's runBare method to
ReplicationRun. This methods first calls BaseTestCase's version, and
then additionally saves away the log and db files for the two common
replication directories sb_slave and db_master.
Injecting a bug in a replication test, I see this subdirectory in failure dir:
../fail/Embedded_40/ReplicationRun_Local_3_p4/testReplication_Local_3_p4_StateNegativeTests:
:
-rw-r--r-- 1 dw136774 66797 Jun 12 20:26 db_master-derby.log
drwxr-xr-x 4 dw136774 512 Jun 12 20:26 db_master-wombat
-rw-r--r-- 1 dw136774 12863 Jun 12 20:26 db_slave-derby.log
drwxr-xr-x 4 dw136774 512 Jun 12 20:26 db_slave-wombat
-rw-r--r-- 1 dw136774 2250 Jun 12 20:26 error-stacktrace.out
so it seems to work as expected.
> Make replication tests save derby.log and database when a failure occurs.
> -------------------------------------------------------------------------
>
> Key: DERBY-4270
> URL: https://issues.apache.org/jira/browse/DERBY-4270
> Project: Derby
> Issue Type: Improvement
> Components: Replication, Test
> Reporter: Dag H. Wanvik
> Assignee: Dag H. Wanvik
> Priority: Minor
> Fix For: 10.6.0.0
>
> Attachments: derby-4270.diff
>
>
> Currently, derby.log is saved when a failure happens for most JUnit test.
> Replication tests use
> special subdirectories, so derby.log and the database files are not saved in
> the failure directory for these tests.
> It would be good to fix this to help diagnose intermittent errors.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
