Skip to content

Commit

Permalink
Remove obsolete properties
Browse files Browse the repository at this point in the history
  • Loading branch information
emmartins committed Jan 14, 2025
1 parent 6c97e68 commit 7bc21b2
Show file tree
Hide file tree
Showing 2 changed files with 1 addition and 3 deletions.
2 changes: 1 addition & 1 deletion jta-crash-rec/README-source.adoc
Original file line number Diff line number Diff line change
Expand Up @@ -169,7 +169,7 @@ WARN [com.arjuna.ats.jta] (Periodic Recovery) ARJUNA016037: Could not find new
WARN [com.arjuna.ats.jta] (Periodic Recovery) ARJUNA016038: No XAResource to recover < formatId=131077, gtrid_length=29, bqual_length=36, tx_uid=0:ffff7f000001:1040a11d:534ede43:1c, node_name=1, branch_uid=0:ffff7f000001:1040a11d:534ede43:20, subordinatenodename=null, eis_name=java:jboss/datasources/JTACrashRecQuickstartDS >
----
+
This is normal. What actually happened is that the first resource, `JTACrashRecQuickstartDS`, committed before the {productName} server was halted to insert the recovery record. The transaction logs are only updated/deleted after the outcome of the transaction is determined. If the transaction manager did update the log as each participant (database and JMS queue) completed then throughput would suffer. Notice you do not get a similar message for the JMS resource since that is the resource that recovered and the log record was updated to reflect this change. You need to manually remove the record for the first participant if you know which one is which or, if you are using the community version of the ${product.name} server, then you can also inspect the transaction logs using a JMX browser. For the demo it is simplest to delete the records from the file system, however, *be wary of doing this on a production system*.
This is normal. What actually happened is that the first resource, `JTACrashRecQuickstartDS`, committed before the {productName} server was halted to insert the recovery record. The transaction logs are only updated/deleted after the outcome of the transaction is determined. If the transaction manager did update the log as each participant (database and JMS queue) completed then throughput would suffer. Notice you do not get a similar message for the JMS resource since that is the resource that recovered and the log record was updated to reflect this change. You need to manually remove the record for the first participant if you know which one is which or, if you are using the community version of the ${productName} server, then you can also inspect the transaction logs using a JMX browser. For the demo it is simplest to delete the records from the file system, however, *be wary of doing this on a production system*.

. Do NOT forget to link:{configureBytemanDisableDocUrl}[Disable the Byteman script] by restoring the backup server configuration file. The Byteman rule must be removed to ensure that your application server will be able to commit 2PC transactions!

Expand Down
2 changes: 0 additions & 2 deletions pom.xml
Original file line number Diff line number Diff line change
Expand Up @@ -48,8 +48,6 @@

<properties>
<root.dir>${project.basedir}</root.dir>
<jboss.home.name>WILDFLY_HOME</jboss.home.name>
<product.name>WildFly</product.name>
<!-- A base list of dependency and plug-in version used in the various quick starts. -->
<version.org.asciidoctor.asciidoctor-maven-plugin>2.1.0</version.org.asciidoctor.asciidoctor-maven-plugin>
<version.org.wildfly.checkstyle-config>1.0.7.Final</version.org.wildfly.checkstyle-config>
Expand Down

0 comments on commit 7bc21b2

Please sign in to comment.