Question
Last activity: 6 May 2017 16:41 EDT
Not able to connect to PegaData in split schema
Hi,
We have upgraded the PRPC 6.2 SP2 to PEGA 7.2 .
We observer that the system is not able to connect the PegaData in split schema. Always It is trying to connect with the Rules schema name to tables that are in the Data schema.
can you please advice what could be the issue?
-
Like (0)
-
Share this page Facebook Twitter LinkedIn Email Copying... Copied!
Pegasystems Inc.
IN
Hi Uthaya,
Which application server you are using?
Can you please share complete startup logs?
We are using websphere8, Can you please let me know your email ID will send you the logs.
Pegasystems Inc.
IN
Its better to share logs here so that people can review and comment.
Did you create Creating JNDI binding identifiers in Websphere? ( kindly refer page-42 of PRPC 7.1.9 oracle-Websphere installation guide)
Ok. The startup log is attached in POST.
Yes we created JNDI binding identifiers for PEGARules and PEGAData schema.
The data source is existing data source which is pointed to PEGARules schema.
Pegasystems Inc.
IN
Got the problem.
You need to connect your Data Source to PegaDATA schema.
Please re-configure as above and restart server.
Share your observation.
Yes when we pointed the existing PEGARules data source to data schema, the system didn't boot up . we got the attached error.
Pegasystems Inc.
IN
Can you please add following in prconfig.xml?
<env name="Identification/SystemName" value="prpc"/>
This required server restart
Virtusa Corporation
US
Hi, we are still getting the same boot up error when we update the above mentioned config in prconfig.xml. Also the system name already given in the DSS as well .
Please advice what could be the issue, We are facing this when we perform the follwing in websphere for the PEGARules datasource in custom property. Here in place of data schema name if we give Rules schema name , The system is booting up but the tables are not pointing to data schema. Please advice.
Name: currentFunctionPath Value: SYSIBM, SYSFUN,<data schema name>
Note: SYSIBM and SYSFUN are system schemas provided by the database. They are required
because the Pega 7 Platform uses some functions made available in these schemas.
l Name: currentSchema Value: <data schema name>
Pegasystems Inc.
US
From the original startup logs we see that it is not picking up the namespace bindings for the default schemas are not getting pick up correctly. What scope did you define them at? Please provide screenshots so we can check to see if there is any spelling issue.
For the currentFunctionPath you should be using the data schema name but it isn't your issue. If you don't have the defaultSchemas setup correctly then nothing is going to get schema qualified correctly.
Virtusa Corporation
US
Thanks for the response. Please find the atatched screenshot for the name binding identifier configuration.
It is specified at node level. Kndly check and advice. Let me knwo if you need any further information.
Pegasystems Inc.
IN
Hi Uthayakumar,
Use server scope for all and then restart server.
Update the outcome.
Virtusa Corporation
US
Hi Aravind,
Which mean do you want me to define name binding at server level scope?
ie . for us Node=pegd101bvaaNode01, server=server1
Pegasystems Inc.
IN
yes, this is what I meant to have.
Virtusa Corporation
US
Also please note that the datasource also currently specified at node level.
Virtusa Corporation
US
Hi Aravind,
Can you confirm , do we need to introduce both datasources and name binding varaible at server level ?
Also just FYI, Please find the prconfig.xml values. Let us knwo if anything is missing.
<env name="initialization/settingsource" value="merged" />
<env name="database/databases/PegaRULES/dataSource" value="java:comp/env/jdbc/PegaRULES"/>
<env name="database/databases/PegaDATA/dataSource" value="java:comp/env/jdbc/PegaRULES"/>
<env name="Initialization/usenativelibrary" value="true" />
<env name="Initialization/explicitTempDir" value="${pega.tmpdir}" />
Pegasystems Inc.
US
I have defined at the server and cluster levels before so I know they both work. Since you are having issue at the node level I would try server as previously suggested.
Virtusa Corporation
US
Hi Celeste , This is an exisitng system and all the datasources realted to this applciation configured at node level. Is possible can you anybody from PEGA system walkthrough the configuration through call ? Moving PegaRules datasource alone to server level scope might impact the system functionality.
Pegasystems Inc.
US
I was actually talking about moving the bindings to the server level. Is this a cluster or standalone system you are trying to start? Are there other pega systems on this server which would conflict with having the bindings at the server level? If so you should open an SR so we can look at it. I don't know that we couldn't get node level bindings to work. I just have never done it at that level.
Pegasystems Inc.
US
SUMMARY
During upgrade from PRPC 6.2SP2 to split-schema configuration, application is not able to connect to the PegaData. Upon binding Data Source to the PegaData schema, the system does not boot. Namespace bindings are configured at node level for Websphere 8 Node
SUMMARY
During upgrade from PRPC 6.2SP2 to split-schema configuration, application is not able to connect to the PegaData. Upon binding Data Source to the PegaData schema, the system does not boot. Namespace bindings are configured at node level for Websphere 8 Node
ERROR MESSAGES
[7/6/16 14:14:55:134 EDT] 00000051 PRBootstrapDa Z com.pega.pegarules.internal.bootstrap.PRBootstrapDataSource Loading bootstrap properties from /prbootstrap.properties
[7/6/16 14:14:55:136 EDT] 00000051 SettingReader Z com.pega.pegarules.internal.bootstrap.SettingReaderJNDI Could not find java:comp/env/prbootstrap/ in the local JNDI context, skipping prconfig setting lookup
[7/6/16 14:14:55:137 EDT] 00000051 SettingReader Z com.pega.pegarules.internal.bootstrap.SettingReaderJNDI Could not find prbootstrap in the local JNDI context, skipping prconfig setting lookup
[7/6/16 14:14:55:137 EDT] 00000051 SettingReader Z com.pega.pegarules.internal.bootstrap.SettingReaderJNDI Could not find java:comp/env/prconfig/ in the local JNDI context, skipping prconfig setting lookup
[7/6/16 14:14:55:138 EDT] 00000051 SettingReader Z com.pega.pegarules.internal.bootstrap.SettingReaderJNDI Could not find prconfig in the local JNDI context, skipping prconfig setting lookup
[7/6/16 14:14:55:138 EDT] 00000051 PRBootstrapDa Z com.pega.pegarules.internal.bootstrap.PRBootstrapDataSource Bootstrap datatables schema: null
[7/6/16 14:14:56:178 EDT] 00000051 PRBootstrap Z com.pega.pegarules.internal.bootstrap.PRBootstrap Engine schema defined in allclasses: null
[7/6/16 14:14:56:179 EDT] 00000051 SettingReader Z com.pega.pegarules.internal.bootstrap.SettingReaderJNDI Could not find java:comp/env/prconfig/ in the local JNDI context, skipping prconfig setting lookup
[7/6/16 14:14:56:179 EDT] 00000051 SettingReader Z com.pega.pegarules.internal.bootstrap.SettingReaderJNDI Could not find prconfig in the local JNDI context, skipping prconfig setting lookup
[7/6/16 14:14:56:180 EDT] 00000051 PRBootstrap Z com.pega.pegarules.internal.bootstrap.PRBootstrap Engine classes schema: null
[7/6/16 14:14:56:180 EDT] 00000051 PRBootstrap Z com.pega.pegarules.internal.bootstrap.PRBootstrap Assembled classes schema: null
STEPS TO REPRODUCE
1. Configure PRPC SplitSchema Namespace bindings at IBM Websphere "Node" or "Cluster" scope.
ROOT CAUSE
Refer to the product documentation: https://docs-previous.pega.com/pega-72-platform-installation-guide-websphere-and-db2-luw.
The spilt schema namespace bindings are looked up from PRPC engine code using a local java JNDI context (Server roots partition) and exisiting lookup code does not handle lookup of namespace bindings bound to Node persistent or Cell persistent JNDI partitions.
RESOLUTION
Make the following change to the operating environment:
Delete the PegaRULES and PegaDATA namespace bindings from "Node" scope and re-create the same bindings at ther "Server" scope. If you have more that one PRPC Node or servers, you will need to re-create the same bindings for each of the servers. Once the bindings successfully defined at the "Server" scope please restart each of the servers. This will help you get the PRPC application started with the spilt schema configuration.
Vantiv
US
Hi ,
Not sure if this is already addressed, we have recently migrated from 6.3 to 7.2.2 on WAS 8.5.5 and had the same issue. All our data source configurations, and JNDI setting are done at Cluster level and when we added the below setting in the prconfig we are able to see the PegaDATA schema mapping correctly during the start up
<env name="initialization/settingsource" value="merged" />
<env name="database/databases/PegaRULES/dataSource" value="java:comp/env/jdbc/PegaRULES"/>
<env name="database/databases/PegaDATA/dataSource" value="java:comp/env/jdbc/PegaRULES"/>
Note : what we noticed is even without this we are able to access the Data schema but somehow the startup log is always showing only the PegaRULES schema name as defaulted.
Thanks,
PGBNRAO