[midPoint] DOCKERED MIDPONT (2 NODES) RUNNING ON AN ORACLE CLUSTER

[Carlos Ferreira]

Hi,

1. I have downloaded the evolveum/midpoint image from docker hub;

2. I have create 2 containers, each one running MIDPOINT in 2 separate
servers;

3. I configured the Config.xml file (in both nodes):

*********************** /var/opt/midpoint/config.xml
****************************************************

(...)

<configuration>
    <midpoint>
        <webApplication>
            <importFolder>${midpoint.home}/import</importFolder>
        </webApplication>
    <repository>

<repositoryServiceFactoryClass>com.evolveum.midpoint.repo.sql.SqlRepositoryFactory</repositoryServiceFactoryClass>
      <baseDir>${midpoint.home}</baseDir>
      <embedded>false</embedded>
      <asServer>true</asServer>
      <database>oracle</database>
      <jdbcUsername>midpoint_wi</jdbcUsername>
      <jdbcPassword>secret</jdbcPassword>
      <jdbcUrl>jdbc:oracle:thin:@(DESCRIPTION =(ADDRESS = (PROTOCOL =
TCP)(HOST = orarac.trt)(PORT = 1521)) (LOAD_BALANCE = yes) (CONNECT_DATA =
(SERVER = DEDICATED) (SERVICE_NAME = prod.trt3.jus.br)))</jdbcUrl>
    </repository>
<taskManager>
  <clustered>true</clustered>
  <jmxUsername>midpoint</jmxUsername>
  <jmxPassword>secret</jmxPassword>
</taskManager>

(...)
*********************** /var/opt/midpoint/config.xml
****************************************************


----->>> midpoint was configured to access an ORACLE DATABASE running in a
CLUSTER;


4. The setenv.sh (/usr/local/tomcat/bin/setenv.sh) file has been configured
as follows:

  a) on node A
CATALINA_OPTS="-Dmidpoint.nodeId=NodeA -Dmidpoint.home=/var/opt/midpoint/
-Dcom.sun.management.jmxremote=true -Dmidpoint.jmxHostName=10.3.190.47
-Dcom.sun.management.jmx
remote.port=20001 -Dcom.sun.management.jmxremote.ssl=false
-Dcom.sun.management.jmxremote.password.file=/var/opt/midpoint/jmxremote.password
-Dcom.sun.management.jmxremo
te.access.file=/var/opt/midpoint/jmxremote.access"

  b) on node B
CATALINA_OPTS="-Dmidpoint.nodeId=NodeB -Dmidpoint.home=/var/opt/midpoint/
-Dcom.sun.management.jmxremote=true -Dmidpoint.jmxHostName=10.3.190.79
-Dcom.sun.management.jmx
remote.port=20002 -Dcom.sun.management.jmxremote.ssl=false
-Dcom.sun.management.jmxremote.password.file=/var/opt/midpoint/jmxremote.password
-Dcom.sun.management.jmxremo
te.access.file=/var/opt/midpoint/jmxremote.access"


3. I have run the script that creates on necessary objects (tables, etc) in
the oracle database;

4. When I start the first node (A, for example), I am able to log in;

5. When I try to login in in the second node (B), I receive the following
message:

" Currently we are unable to process your request. Kindly try again later."

6. In the "idm.log" file, I have the message:

2017-10-24 19:35:05,771 []
[QuartzScheduler_midPointScheduler-NodeB_ClusterManager] WARN
(org.quartz.impl.jdbcjobstore.JobStoreTX): This scheduler instance (NodeB)
is still active but was recovered by another instance in the cluster.  This
may cause inconsistent behavior.

7. If I drop all the objects and re-execute the install script and try to
login firstly from node B, I am successful. Nevertheless, I receive the
same error message when trying to log in from node A.


Did I miss anything?


Thks,

Carlos A Ferreira
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.evolveum.com/pipermail/midpoint/attachments/20171024/891d41e2/attachment.htm>