(20223247) Direct SQL deployment fails on an Oracle database due to a bug related to database links.
If you are using direct SQL deployment on an Oracle database and your publishing and production database schemas are on different machines, you may run into errors caused by an issue with the database links in certain versions of Oracle Database.
Workaround: See the following document for information about this issue and how to resolve it:
https://support.oracle.com/rs?type=doc&id=18759589.8
(21459580) Performance issues when importing data into versioned repositories on Microsoft SQL Server.
When importing data into versioned repositories on a Microsoft SQL Server database, performance may be slow due to full table scans.
Workaround: To prevent the table scans, create separate indexes for the asset_version
column and the sec_asset_version
column in each table in which the scans are occurring.
(14248657) On OJBDC drivers, errors may occur when using SqlPassthroughQuery
bind variables syntax.
When creating a service that uses the SqlPassthroughQuery
bind variables, using the following syntax will create an error:
select * from dcspp_order where ?0 = 1
Workaround: Use one of the following instead:
select * from dcspp_order where ? = 1
select * from dcspp_order where :0 = 1
(14037115) If you are using a MySQL database for a development installation, your initial deployment may fail. This failure may be caused by the MySQLIntegrityConstraintViolationException
exception.
Workaround: Configure your server data sources to use the com.mysql.jdbc.Driver
driver class and use the EmulateTwoPhaseCommit
global transaction protocol. The following example Java Database Connectivity (JDBC) XML file makes these configurations.
<jdbc-data-source> ... <jdbc-driver-params> ... <driver-name>com.mysql.jdbc.Driver</driver-name> ... </jdbc-driver-params> <jdbc-data-source-params> ... <global-transactions-protocol>EmulateTwoPhaseCommit </global-transactions-protocol> </jdbc-data-source-params> </jdbc-data-source>
(13469587) Reference Update Phase in deployment runs slowly on an Oracle database
Inefficient queries can occur when performing a deployment on an Oracle database. Analysis shows a SQL query with sub-queries that run for each SKU in the product catalog, taking a long time to complete.
Workaround: Set the useEagerLoadingStrategyForSourceItemPropertyValue
property on the /atg/deployment/DeploymentManager
component to true
and the useUnnestHintForSELECTMAXInnerSQLQuery
property on the /atg/epub/version/VersionManagerService
component to true
. Refer to article 1416520.1 in the My Oracle Support knowledge base for more information.
(13295756) Deployment failures occur if you use ATG Content Administration to delete sites.
It is possible to delete sites through the ATG Content Administration user interface. However, doing so is not supported (in any UI). If you attempt to deploy a project containing deleted sites, deployment fails and SQL Integrity Constraint errors may occur.
Workaround: Disable the sites through Site Administration rather than deleting them.
(13284078) Deployment of standalone EAR file on an IBM WebSphere asset management server strips VFS version numbers from filenames.
Workaround: Copy the versionFileStore
directory that contains all filenames and version numbers to the WebSphere’s deployed EAR file on the WAS server location.
(13283769) In the Business Control Center, attempts to add a site returns user to existing site configuration.
This problem occurs when navigating as follows:
Visit an existing target site’s configuration.
Return to Deployment Administration Configuration and click Add Site. Doing so brings up the previously-viewed site configuration instead of an empty form.
Workaround: Before trying to add a site, do one of the following:
From the existing site’s configuration page, click on the link Back to Deployment Administration Configuration.
From Deployment Administration Overview, click Refresh Results.
(13277154) While a one-off deployment is in progress, attempts to perform any action within the project where the one-off deployment was initiated yield an error message .
Workaround: Wait until the one-off deployment is complete.
(13276781) When you save a new or modified scenario in the ACC, the folder asset /atg
is added to the project along with the scenario files.
Workaround: Remove the folder asset from the project.
(13276779) When you add an existing scenario to a new project from the ACC, the scenario files are not added to the project. Instead the root directory ‘/atg’ is added as a fileFolder
asset.
Workaround: Change the scenario in the ACC, then save it.
(13276524) Unable to recover deployment that fails due to file ownership conflict.
(13275484, 13277610) DB2 deadlocks occur during deployments over multiple Content Administration threads.
This problem is due to IBM’s locking strategy (see PMR #42831,756,000), which constrains ATG deployments to a single thread.
Workaround: Deploy from Content Administration via a single thread/server. Set the following DeploymentManager
component properties (/atg/deployment/DeploymentManager
):
maxThreads=1
useDistributedDeployment=false
(13274544) A project’s Assets page does not display conflicts between properties whose expert attribute is set to true
.
Workaround: Set the property’s expert attribute to false
.
(13274142) ATG Content Administration allows users to advance the workflow for an asset after its parent folder is deleted by another project.
The Assets page incorrectly displays the asset’s path as root (/
).
(13274112) Preview of session-scoped slots show changes only once within a given session.
Workaround: In order to preview changes more than once within the same project, temporarily set the slot’s scope to request; reset the scope to session when it is ready for deployment.
(13274106) Errors occur when you deploy personalization assets (such as slots) whose scope is not set to global—for example, request or session.
The deployment executes as expected, and these error messages can be safely ignored.
(PUBLISHING-131860) Running more than one instance of ATG Content Administration on the same machine causes port conflicts.
If your environment includes a cluster of ATG Content Administration servers on the same machine, the following error appears when you start the second server:
Service /atg/epub/file/synchronization/
FileSynchronizationDeployServer found port 8815 is already in use.
Workaround: For each additional ATG Content Administration server, create an /atg/epub/file/synchronization/FileSynchronizationDeployServer.properties
file and change the localport
property value from 8815 to any unused port number. Make sure the new file is in your ATG Content Administration config
layer.
Note that changing the port through the /atg/dynamo/Configuration.properties
file does not work correctly in this situation.