Limitations

Be aware of the following limitations:

• IBM Cloud Orchestrator V2.4 can be used to manage IBM PowerVM infrastructure through the IBM PowerVC virtualization platforms only in development or test environments.
• When managing a VMware environment, datastore names should not have a space. Otherwise, deployments will fail attaching config_drive to server.
• You cannot install more than one IBM Cloud Orchestrator V2.4 instance on the same vCenter environment if the instances have access to the same resources. Each IBM Cloud Orchestrator V2.4 instance must use a different user ID to access the vCenter. The intersection between the resources that are seen by these users (for example, clusters, datastore) must be empty.
• The DataBasePassword parameter is included in some Heat templates, but this parameter is not supported. Do not use this parameter.
• When you add the Default add disk add-on to SoftLayer images in Virtual System (Classic) Patterns, ensure that you request disk sizes that match the available disk options in SoftLayer. Otherwise, the add-on will fail to format and mount the disk when deploying the pattern.

Known Issues

General Issues | Installation and Upgrade Issues | IBM Workload Deployer Issues | User Interface Issues | Documentation Issues
General

• If a project is deleted, all of its assigned resources (virtual machines, stacks, patterns, networks, images, and so on) remain in the cloud. Only the Cloud Administrator can manage these orphan resources. The Domain Administrator cannot recover from this situation.
• When scaling your IBM Cloud Orchestrator environment by adding nodes manually, you must add one node at a time. This approach ensures that the node is created, unless an OpenStack quota is exceeded or other limitations prevent the node creation.
• When you deploy an instance to a VMware region, the name that is specified at deployment time is propagated from the Self-service user interface to the OpenStack layer. The instance is created in the VMware hypervisor by using the corresponding OpenStack UUID. To match an instance in OpenStack with the actual instance deployed on VMware, complete the following steps:
1. On the Region Server where the instance was deployed: Run the nova list command, and identify the instance UUID in the command output.
2. In the vCenter client: In the Search field, type the instance UUID, and press Enter.
• If the SCOrchestrator.py script prompts for a password when accessing the Region Server, edit the ˜/.ssh/authorized_keys file to check whether two keys are joined together, as shown in the following example:
  
  == root@<some-host-name>ssh-rsa
  
  If you find such entries, insert a new line before the ssh-rsa text. The ssh-rsa text should always be at the beginning of a new line.
• If a virtual machine fails to deploy, or if you cannot access the virtual machine, the local /etc/hosts file on the Region Server might be incorrectly configured. Edit the /etc/hosts file to check whether two host entries are on the same line, as shown in the following example:
  
  10.0.0.1 host110.0.0.2 host2
  
  If you find such entries, insert a new line before the second IP address, and validate the file format.
• If a virtual machine with the Default Add Disk add-on is deployed to a SoftLayer, z/VM, or Amazon EC2 region, the attached volume might not be detected at the operating-system level. The log file contains the following text after the add-on is run:
  
  Warning: WARNING: the kernel failed to re-read the partition table on /dev/vdc (Device or resource busy).
  
  As a result, the add-on is in the error state, and the virtual machine might not reflect all of your changes until after reboot.

• During installation and upgrade, IBM Cloud Orchestrator V2.4 passwords can contain only alphanumeric characters and underscores. Spaces, hyphens, and other special characters are not allowed.
• Before you upgrade from IBM SmartCloud Orchestrator V2.3.0.x to IBM Cloud Orchestrator V2.4.0.0, ensure that the passwords do not include any special characters, or the upgrade is likely to fail. For instructions about how to change the passwords before the upgrade, see Changing the various passwords (V2.3). It is not necessary to remove the special characters from operating system (OS) passwords.
• During the installation or upgrade, specify passwords that do not contain any special characters.
• After the installation or upgrade, change the passwords to include special characters in accordance with security best practices. For instructions about how to change the passwords after installation or upgrade, see Changing the various passwords (V2.4.0.1).
• Before you upgrade from IBM SmartCloud Orchestrator V2.3.0.x to IBM Cloud Orchestrator V2.4, ensure that VMware datastore names do not contain any spaces.
• After you upgrade to IBM Cloud Orchestrator V2.4, if you use the same browser session as before the upgrade, the PATTERNS submenus are not displayed correctly. To correct this problem, clear your browser cache and restart your browser.
• After you upgrade to IBM Cloud Orchestrator V2.4, deploying a Virtual System Pattern (classic) fails with the following error:
  
  Could not contact virtual machine over the network to verify initialization.
  
  This problem occurs because the V2.3 images were not mapped correctly into V2.4 format during the upgrade.
  
  To recover from this situation, run the following command on one line on Central Server 3:
  
  python /var/chef/cache/cookbooks/workload_deployer/files/default/restore_image_mappings.py -u admin -p admin_password -i central_server_3_hostname -f /drouter/ramdisk2/mnt/raid-volume/raid0/templates/rainmaker-templates/image_mappings_backup.txt
• After you upgrade from SmartCloud Orchestrator V2.3, the IBM Cloud Orchestrator V2.4 user interfaces no longer run on Central Server 3. Instead, the user interfaces are accessible through the IBM HTTP Server, which runs on Central Server 4. For more information, see Accessing IBM Cloud Orchestrator user interfaces.
• After you upgrade to IBM Cloud Orchestrator V2.4, complete any necessary post-upgrade configuration tasks as described in Performing post-installation tasks and Performing advanced configuration tasks.
• After you upgrade to IBM Cloud Orchestrator V2.4, VMware discovery is not started and is not configured automatically. If you would like to use VMware discovery, you must configure and start it as described in Configuring vmware-discovery and Configuring vmware-discovery for multiple vCenters or cluster/resource pool.

• Virtual System Patterns do not support flavors with a disk size of 0, regardless of the type of hypervisor used. For Virtual System Pattern deployments, use a flavor with a disk size that is equal to or greater than the disk size of the image.
• After you schedule a pattern for deployment at a future time, the Instances page for the pattern displays a blank page with an error message. You can disregard this message, and wait for the deployment operation to complete. If you want to delete the scheduled pattern deployment, use the deployer.virtualapplications.delete(depl_id) Workload Deployer Command Line Interface (CLI) command to delete the appropriate virtual pattern instance.
• It is not possible to import multiple pattern types into IBM Cloud Orchestrator at the same time. Multiple pattern types must be imported one at a time: the next import can start only if the previous imported pattern is visible in the list of available pattern types.
• When you create a Linux image with the cloud-init activation technology for use in Virtual System Patterns, those images should have IPv6 networking disabled.

To disable IPv6 networking in Red Hat Enterprise Linux, complete the following steps:

1. Edit the /etc/sysconfig/network file to set the IPV6_AUTOCONF parameter to no:
   IPV6_AUTOCONF=no
2. Restart networking, as follows:
   service network restart

To disable IPv6 networking in SUSE Linux Enterprise Server, complete the following steps::
1. Add the following lines to the /etc/sysctl.conf file:
   net.ipv6.conf.default.autoconf = 0
   net.ipv6.conf.all.autoconf = 0
   net.ipv6.conf.eth0.autoconf = 0
2. Load the kernel parameters from the /etc/sysctl.conf file, as follows:
   sysctl -p
3. Restart networking, as follows:
   service network restart
• Microsoft Windows virtual machines only: If you click PATTERNS > Instances > Virtual System Instances, select a running instance, and click Manage, an error is displayed instead of the management console page if the instance has a Windows virtual machine.
• IBM WebSphere® Virtual System Pattern templates must be migrated to the version of IBM OS Image for Red Hat Linux Systems that is installed in your IBM Cloud Orchestrator environment. To migrate a template, complete the following steps:
1. Click PATTERNS > Pattern Design > Virtual System Templates.
2. Open the virtual system pattern template that you want to migrate.
3. In the Pattern Designer, you are prompted to select alternative image types for the images that are referenced by this pattern, but do not exist in this environment.
1. From the Interchangeable image types list, select the appropriate image.
2. Click OK to change the image type referenced by the virtual system pattern template.
4. Save the template.
• When you use Linux images with the cloud-init activation mechanism to deploy Virtual System Patterns, the deployment process might result in error, with the following information in the /0config/0config.log log file on the deployed virtual machine:
  
  [date and time stamp] maestro Successfully downloaded file using pipe
  [date and time stamp] install_im.py /hone/virtuser exist:True
  [date and time stamp] invoker An error occurred; see trace.log for details.
  [date and time stamp] invoker Traceback (most recent call last):
  File "/0config/nodepkgs/common/python/invoker.py", line 227, in execute
  execfile(pyfile, _pyglobals)
  File "install_im.py", line 60, in <module>
  grp = grp.getgrnam('virtuser')
  KeyError: 'getgrnam(): name not found: virtuser'
  
  To avoid this problem, when you prepare a virtual image for use with Virtual System Patterns, create the virtuser user as a member of the virtuser group in the operating system.
• If you are using the Pattern Builder to view the list of flavors for a particular image, and you add a new flavor, you cannot see the new flavor for this image until you refresh the Pattern Builder page.
• When you use Linux images with the cloud-init activation mechanism to deploy Virtual System Patterns, the deployment process might result in the following errors:
  
  [date and time stamp] 00000001 VariableExpan E WSVR0244E: An undefined HOST product variable has been encountered in the krb5Spn property of the /opt/IBM/WebSphere/Profiles/DefaultAppSrv01/config/cells/CloudBurstCell_11411667517672/security.xml#KRB5_1 configuration object.
  [date and time stamp] 00000001 WsServerImpl E WSVR0100W: An error occurred initializing, server1 [class com.ibm.ws.runtime.component.ServerImpl] com.ibm.ws.exception.ConfigurationError: com.ibm.wsspi.runtime.variable.UndefinedVariableException: Undefined variable HOST
  
  To avoid this problem, complete one of the following steps:
• For the base operating-system image, use the IBM OS Image for Red Hat Linux Systems, which you can purchase from IBM if available for your IBM Cloud Orchestrator environment.
• Contact IBM Software Support to get a custom script package that configures the necessary settings. This script package must be the first package that runs when the virtual machine starts, so that other middleware software can start in the correct environment.

• The Administration user interface sometimes fails with an Internal Server Error, and the associated log file (/var/log/httpd/openstack-dashboard-error.log) contains the following entries:


[date and time stamp] [error] [client IP_address] if self.is_usable():
[date and time stamp] [error] [client IP_address] File "/usr/lib/python2.6/site-packages/ibm_db_django/base.py", line 240, in is_usable
[date and time stamp] [error] [client IP_address] if self.databaseWrapper.is_active(connection):
[date and time stamp] [error] [client IP_address] NameError: global name 'connection' is not defined

To resolve this problem, complete the following steps:

1. Update the /usr/lib/python2.6/site-packages/ibm_db_django/base.py file as follows:
   Before: if self.databaseWrapper.is_active(connection):
   After: if self.databaseWrapper.is_active(self.connection):
2. Restart the Administration user interface.
• Microsoft Internet Explorer 10 browser only: When you use the Pattern Builder for virtual systems, the flavors list is not displayed correctly. The number of displayed flavors is reduced, and the navigation bar is missing at the bottom of the window.
• Microsoft Windows virtual machines only: When you use the Self-service user interface to deploy a Windows virtual machine, by clicking SELF-SERVICE CATALOG > Deploy cloud services > Deploy a single virtual server, do not edit the UserId field. You can change the password only for the user who is specified in the cloudbase-init configuration file.
• In the Administration user interface, when you click Project > Instances, it is not possible to scroll backwards through multiple pages of instances. To return to a previous page, click Project > Instances in the left navigation pane, and navigate forwards to the page of choice.
• Default domain quota values are not created for new domains: in the Self-service user interface, if you click CONFIGURATION > Domain > new-domain > Quota, no values are displayed. To resolve this problem, use the Administration user interface to edit the domain quotas as described in Editing the domain quotas, and click Save.
• Public Cloud Gateway only: After editing the configuration information for Public Cloud Gateway regions in the config.json file, you must run the following command on the Public Cloud Gateway server, to propagate the changes to the rest of the system:
  
  refreshEndpoint.sh admin admin_password `hostname -f`:9443
  
  If the command was successful, the command output includes the following line:
  
  HTTP/1.1 204 No Content
  
  To see any new Regions or Availability Zones in the user interface lists, users must log out of the user interface and log back in again.
• In the Administration user interface, click ADMIN > System Panel > Images. Select one of the default images provided with IBM Cloud Orchestrator, and click Edit. Expand the Configuration Strategy section. From the Template Source list, select Local File, and click Browse. When you click Open, the following error is displayed:
  
  There was an error parsing the configuration strategy JSON file. Make sure the file contains well formed JSON.


Workaround: If the configuration strategy is initially applied to an image using the Administration user interface, then the Administration user interface must be used to export and re-apply the strategy to other images. If the configuration strategy is initially applied using the Command-Line Interface (CLI), then the CLI must be used to export and re-apply the strategy to other images. For the default images that are provided with IBM Cloud Orchestrator, the configuration strategy is applied through the CLI, so you must use the OpenStack Glance CLI to export and import the strategy.

• Microsoft Internet Explorer 10 browser only: To avoid potential rendering issues in the Administration user interface, update your browser to the latest patch version. Internet Explorer 10.0.9200.17089 works as expected.

• In the "Setting up virtual machines" topic, in the "Creating the virtual machines" section, update step 5 regarding the contents of the <SERVER_NAME>.xml file, as follows:
  
  Before: <clock offset='localtime'/>
  After: <clock offset='utc'/>
• OpenStack Ceilometer corrections:
• The Verifying the installation topic indicates that Ceilometer is started by default, which is not true.
• The Region Server automation policy indicates that Ceilometer is part of the policy, which is not true.
• The documentation does not specify the RPM package names for the required python modules:
• Red Hat Enterprise Linux:
  - python
  - python-pycurl
• SUSE Linux Enterprise Server
  - python
  - python-base
  - python-curl
• The documentation does not specify the minimum version of Python required. The minimum version is Python V2.6.2.
• The documentation does not include information about the Workload Deployer command-line interface (CLI). For more information about the Workload Deployer CLI, see Using the command-line interface (V2.3) in the IBM SmartCloud Orchestrator V2.3 documentation. You can download the Workload Deployer CLI for IBM Cloud Orchestrator V2.4 from the following web page, where workload_deployer_host_name is the fully qualified host name of the server where the Workload Deployer component is installed (usually Central Server 3 in a distributed installation topology):
  
  https://workload_deployer_host_name/downloads/cli
• The instructions about how to change the admin, bpm_admin, tw_admin, db2inst1, and OpenStack passwords are incorrect in Changing the various passwords (V2.4). For the correct instructions, see Changing the various passwords (V2.4.0.1).
• The instructions about how to change passwords are incorrect in Changing password (V2.4). For the correct instructions, see Changing password (V2.4.0.1).
• The instructions about how to change the Deployment Service passwords are incorrect in Changing the Deployment Service password (V2.4). For the correct instructions, see Changing the Deployment Service password (V2.4.0.1).
• The documentation does not include instructions about how to use the Deployment Service wizard to upgrade from IBM SmartCloud Orchestrator V2.3.0.x to IBM Cloud Orchestrator V2.4.0.0. For these instructions, see Technote 1686654.
• The documentation does not include instructions about how to switch Workload Deployer database remote access to a different user. For these instructions, see Technote 1685992.

To access the latest Technotes, search the IBM Techdocs library.




Back to top

Open defects

Review the following list of open defects for IBM Cloud Orchestrator on the IBM Support Portal.