CloudBees CD v10.0.3 is a maintenance release (MR) applied to the v10.0.0 long-term release. As such, consult the following in combination with these notes for everything that this release provides.
For more details see CloudBees CD release strategy.
- This release includes the following security updates
The Apache web server used in CloudBees CD is upgraded from v2.4.46 to 22.214.171.124. For details, see https://www.apache.org/dist/httpd/CHANGES_2.4. [NMB-29873]
PHP is upgraded from v7.4.12 to 126.96.36.199. For details, see https://www.php.net/releases/7_4_13.php. [NMB-29873]
OpenSSL is upgraded from 1.1.1h to 1.1.1i. For details, see https://www.openssl.org/news/openssl-1.1.1-notes.html. [NMB-29873]
- General features
- New platform support
- Database support
- Browser support
Bug in java 11.0.6 leads to connection issues with the EC-OpenShift plugin.
PW01 jobs not processing after 3.4 million deferred messages on Feb15 & splitbrain in jgroups channels. Message Backlog started on Feb9 as jobs stuck after backgroundDelete deletes session in use by recheckSubscriptions.
Job step taking excessive time while doing
In SAML SSO
In some circumstances, slowness detected in automation platform UI because a long batch of getProperties requests was sent twice.
Obsolete modules have been excluded from JRE 11.0.9.
Knowledge base article link updated about using mysql with default latin collation.
Response time for displaying Pipelines Runs list and Releases list UI has been improved. To enable the fix, add
Job steps waiting for a resource on a macOS Catalina CloudBees CD/RO agent were unable to create a directory on a mounted workspace folder.
When using nested pipelines, the dependent POST gate task/rule, which has
When creating a table widget for a DevOps Insight dashboard, the 10th column added was placed in the second spot, which pushed columns 2-9 forward by one spot.
Fixed the SearchGuard permissions for scroll. Multiple hours delay from exponential backoff when processing
Fixed a problem with creating a table widget for a DOIS dashboard: the 10th column added was placed in the 2nd spot which pushes columns 2-9 forward by 1 spot.
- Plugin updates
Addressed a security issue. [NMB-30227]
Dependencies updated. [FLOWPLUGIN-8991]
Problem fixed with plugin issuing the error
For complete installation and upgrade information:
|Before starting an upgrade, make sure to back up your existing CloudBees CD data.|
- Upgradable versions
Upgrades to CloudBees CD 10.x are only supported only from ElectricFlow 8.0 or later.
Upgrades from version 4.2 or earlier are not supported.
For upgrade instructions, see Upgrade roadmap.
- Upgrading traditional deployments
When upgrading a traditional deployment from version 10.0.3 or earlier, the
DATA_DIRECTORY/broker-datadata folder must be deleted prior to the upgrade.
- Upgrading an older DevOps Insight version
Upgrades from version 9.0.x, or earlier: requires upgrading the DevOps Insight server to 10.0.2.
Upgrades from version 9.1, or later: upgrading is not required.
- Updating elements containing applicationServiceMapping
If your XML export file from CloudBees CD 8.0.1 or earlier versions has elements containing
applicationServiceMapping, you must change all instances of that string in the file to
serviceClusterMappingbefore importing the file into version 10.0.3. For example, change the following XML:
<applicationServiceMapping> <applicationServiceMappingId>9efcda31-a85f-11e7-8500-0800279f198d</applicationServiceMappingId> <applicationServiceMappingName>9efcda31-a85f-11e7-8500-0800279f198d</applicationServiceMappingName> … </applicationServiceMapping>
<serviceClusterMapping> <serviceClusterMappingId>9efcda31-a85f-11e7-8500-0800279f198d</serviceClusterMappingId> <serviceClusterMappingName>9efcda31-a85f-11e7-8500-0800279f198d</serviceClusterMappingName> … </serviceClusterMapping>
- Backing up and restoring custom settings
The DevOps Insight installer overwrites the
elasticsearch.ymlconfiguration file with a new file. As of DevOps Insight version 8.3, the file includes a
Custom Settingssection, which lets you add Elasticsearch settings not managed by the DevOps Insight server without being lost during an upgrade. [NMB-25850]
Upgrading from version 8.2, or earlier: if you added settings to this file that you want to preserve, you must back up the file to a separate location before upgrading to version 10.0, or newer, and then add the settings to the
Custom Settingssection after the upgrade.
Upgrading from version 8.3, or later: No action required. The installer preserves the settings in the
- Updating the MySQL configuration before upgrading
Since release 8.0.1, CloudBees has instructed customers using a MySQL database to use the following two lines in their MySQL configuration:
init_connect='SET collation_connection = utf8_unicode_ci, NAMES utf8' skip-character-set-client-handshake
If upgrading from version 8.5, or later, you must remove these lines or comment them out. Otherwise, jobs will not start.
- Installing the MySQL connection jar
CloudBees CD does not include the MySQL JDBC connector. If you plan to perform a CloudBees CD server installation that connects to a MySQL database, you must obtain and install the
MySQL JDBC Connector/J 8.
Sites upgrading from 10.0.* may continue to use
MySQL JDBC Connector/J 5.1 if currently doing so. In which case, no further configuration is required.
See Installing the MySQL JDBC connector for complete details. [NMB-29016]
- Ensuring the correct default MySQL collation
Since release 8.0.1, CloudBees has instructed customers using a MySQL database to ensure that the default collation for their database schema is set to
utf8_general_ciand that no table in their schema overrides this. As of version 9.0, the CloudBees CD server checks this configuration on startup and logs errors in the server log if it is not set correctly.
If the collation is not configured correctly, then entering non-ASCII text into CloudBees CD might cause errors. For example, setting a release name to a non-ASCII value and attempting a search causes an exception.
If your MySQL database schema or any tables in it are set to a non-UTF-8 collation order, see Knowledge Base article KBEC-00385 - Converting a MySQL Database From Latin-1 to UTF-8 for detailed instructions about safely converting your schema to UTF-8. [NMB-26521, NMB_27459]
- Upgrading agents that run the ec-groovy job step in multizone deployments
In multizone CloudBees CD deployments, CloudBees CD agents that are in a different zone than the CloudBees CD server must be upgraded to version 9.0 or later for the
ec-groovyjob step to run successfully on those agents. You must also upgrade the gateway agents that lead back to the server’s zone including those in any zones in between the agent’s zone and the server’s zone. [NMB-27490]
- Removing the SSL 2.0 Client Hello or SSLv2Hello protocol from your security configurations
CloudBees recommends removing the
SSL 2.0 Client Helloor
SSLv2Helloprotocol from your security configurations for all components. [NMB-27934, NMB-29326]
Upgrade agents older that fall into this category for security reasons:
Windows, Linux: 6.0.3 or older; 6.2 or older
Mac OS: 8.4 or older
If this warning appears on the Automation Platform UI:
Note: We recommend removing `SSL 2.0 Client Hello` format from server configuration and upgrade older agents as indicated on the Cloud/Resources Page to avoid security risk.
then enter the following command on the CloudBees CD server:
$ ecconfigure --serverTLSEnabledProtocol=TLSv1.2
- Upgrading the DevOps Insight server
This section provides information about upgrading the DevOps Insight server from Version 7.3 to Version 10.0.2.
- Re-Specifying configuration settings
The installers (GUI, interactive console, and silent mode) for the DevOps Insight server do not preserve the configuration setting for the DevOps Insight server host name (
--hostName) or the setting for the Elasticsearch number of shards (
--elasticsearchNumberOfShards) during the upgrade from 7.3 to 9.2. If you specified non-default values during the 7.3 Reporting server installation, you must re-specify these settings during the upgrade. (All other settings are preserved.)
- Performing a full import
During a full import, the import operation might hang in the following scenarios. To import successfully into CloudBees CD 8.0 and newer versions, perform the appropriate workarounds [CEV-15447, CEV-11873]:
A manual process step in a process has formal parameters. The workaround is to remove the entry related to the property sheet for the job step that is associated with the manual process step.
In the exported XML file from the earlier release, two pipelines are in different projects, and both pipelines have no gate tasks. The flow associated with the pipeline is duplicated under both projects. The workaround is to remove the flow element under the projects.
- Updating application component plugins
Application components are based on plugins (EC-Artifact, EC-Maven and EC-FileSysRepo) with plugin details (such as name, procedure, and parameters) stored in properties on the component. Before version 5.4, the value for the
pluginProjectNameproperty included the plugin name and version (for example, EC-Artifact-188.8.131.52), which closely tied components to specific versions.
When you export your project data before upgrading from ElectricCommander 5.0, 5.1, 5.2, or 5.3 to CloudBees CD 8.x, or later, you must update the application component plugin versions in the export file to the versions on the target CloudBees CD server before importing the data to CloudBees CD 10.0.
For example, if the promoted EC-Artifact plugin version is 184.108.40.206, then in the snippet below, you would change EC-Artifact-220.127.116.11 to EC-Artifact-18.104.22.168:
<property> <propertyId>3f509ffd-506b-11e6-9960-f01faf2c26a3</propertyId> <propertyName>pluginProjectName</propertyName> <counter>0</counter> <createTime>2016-07-23T00:20:20.829Z</createTime> <expandable>1</expandable> <lastModifiedBy>admin</lastModifiedBy> <modifyTime>2016-07-23T00:20:20.829Z</modifyTime> <owner>admin</owner> <tracked>1</tracked> <value>EC-Artifact-22.214.171.124</value> </property> <property>
Starting with release 5.4, CloudBees CD uses the plugin key, which does not include the version, when defining a component. [CEV-6679]
- DevOps Insight server configuration notes
For a production environment, CloudBees recommends that you install the DevOps Insight server on a system other than systems running other CloudBees CD components (such as the CloudBees CD server, web server, repository server, or agent). If you must install it on the same system (such as for testing or other non-production or trial-basis situations) see DevOps Insight server with other components for details.
When an application is cloned from one project (the original project) to another (the destination project), the tier maps for the application will point to the environments with the same names in the destination project. To deploy the application to the environments in the original project, you must create tier maps connecting the application to those environments.
- CloudBees CD on Kubernetes configuration notes
The CloudBees CD server and agent Helm charts provide CloudBees default configuration values found here. Of note, the
images.tagparameter defines the default Helm chart version of record associated with this release. The CloudBees CD/RO
images.tagvalue associated with version 10.0.3 is
Some antivirus software might flag
When you use the Automation Platform UI to upload and publish artifact files with non-English characters in their file names the operation fails with the following error:
Modifications of LDAP user data (such as email addresses) on an Active Directory server after registration in CloudBees CD do not appear properly in user details (in the Automation Platform UI, the Deploy UI, or
(Windows platforms only) If the Elasticsearch cluster which is used by DevOps Insight is in the red state (in Elasticsearch this means that it only partly functions and some data is unavailable) then upgrade reconfigure or uninstall operations will not work. Because the Elasticsearch service can not be stopped when a cluster is in red state kill the Elasticsearch service process by the task manager before running the installer for these actions.
The Microsoft Edge browser does not work with SAML 2.0 and a self-signed certificate during redirection from the identity provider to the service provider. Edge is not recommended for login via SAML 2.0.
Can’t ignore server mismatch and override passkey from Database Configuration page.
The LANG environment variable must be set to
In some circumstances, the StepScheduler becomes deadlocked when processing and scheduling steps.
Error prompts for runtimes started by a schedule are not visible if the schedule was created with a missed configuration.
The stage inclusion status in the Release Dashboard changes color after a stage is renamed.
No error prompt appears for failed tasks and retry tasks during a pipeline runtime.
If an application process step cannot expand to its child steps (because of an invalid run condition or an invalid formal parameter) then the step is not retried even if it uses "retry on error" error handling. The job eventually completes with an error.
The retry count for group tasks or rules using "automated retry on error" is missing from the Pipeline runtime page.
Multiple mapped environments with the same name from different projects are not supported in email notifications.
A project import might not include the path-to-production view.
Jobs might not appear upon drill-down into the "Clusters With Most Deployments" widget in the DevOps Insight Microservices Dashboard if the service does not contain a deploy step in the process.
When you do a full import from version 8.0 to version 8.2 or newer and two or more releases have the same name (under different projects) and are associated to the same pipeline then after import the runs for all releases might become associated to the first imported release. This is because CloudBees CD cannot differentiate runs between the releases since all runs are under the same pipeline project and have the same name. To work around this issue rename releases in the export file so that all their occurrences (in
All subreleases of a release must appear before the release in the DSL for the release-to-subrelease link to be created.
The ability to search by assignee in a Deployment Report is not available in the DevOps Insight report editor.
If Release Command Center was setup for JIRA for user-stories and defects and the JIRA project name was mapped to the release project name using the following field mapping: ` projectName:releaseProjectName` then before upgrading to 10.0 the field mapping must be updated to mention the actual release project name using the following field mapping format:
Long custom labels in email notifications do not render correctly.
Approval by email on manual tasks should not expect parameters.
Navigation to a sub-release editor takes user to the parent release editor. As a workaround, select the subrelease from the left-hand navigation in the parent’s release editor.
When you use the CloudBees CD UI to edit a resource pool and add a tag while renaming it at the same time, the operation fails with the following error:
Running an application process with a parallel manual application process step or running an application process with a parallel manual application and component process steps fails to delete the project.
Attempt to delete a project containing a
Users are not able to delete a project if there are Jenkins builds associated with this project that are references in releases not in the project.
Attempt to delete a build from a pipeline run via
If you use the
These service catalog items are disabled because underlying plugin has been removed.
Step logs local to Windows agents are not visible from UI webserver with
Perform the following on each Windows machine that is a CloudBees CD web server.
You can revert changes only for high-level design objects such as applications procedures procedure steps workflow definitions and state definitions.
Enabling Recursively Traverse Group Hierarchy might impact system performance when the LDAP group hierarchy is traversed. The amount of impact varies with the configurations of the CloudBees CD and LDAP servers the depth of group hierarchy in the LDAP server and the network latency between the servers. Make sure that your directory provider can handle the additional load for supporting nested group hierarchy traversal.
System performance might decrease if you disable change tracking at the server level and then re-enable it. (Change tracking is enabled by default.) For details about using change tracking see change tracking.