Skip Headers
Oracle® Exalogic Elastic Cloud Release Notes
Release EL X2-2, X3-2, and X4-2

E18480-13
Go to Documentation Home
Home
Go to Book List
Book List
Go to Table of Contents
Contents
Go to Feedback page
Contact Us

Go to previous page
Previous
Go to next page
Next
PDF · Mobi · ePub

1 Release Notes for Exalogic Elastic Cloud Software

This chapter lists the new features and known issues in the Exalogic Elastic Cloud Software (EECS).

1.1 New Features and Enhancements in EECS 2.0.6

This section describes the new features and enhancements introduced in release 2.0.6 of the Exalogic Elastic Cloud Software (EECS).

Besides enhanced reliability, robustness, performance, and usability, EECS 2.0.6 includes the following features and enhancements:

1.1.1 Deploying Assemblies in the Exalogic vDC Using OVAB Deployer

When you upgrade to (or install) EECS 2.0.6, OVAB Deployer 11.1.1.6.2 is installed in the Exalogic Control VM. You can use OVAB Deployer to deploy instances of OVAB-generated assemblies in the Exalogic vDC.

For more information, see the "Deploying Assemblies in the Exalogic vDC Using OVAB Deployer" chapter in the Oracle Exalogic Elastic Cloud Administrator's Guide.

1.1.2 Support for Hybrid Configurations

An Exalogic machine in a physical configuration—that is, all the compute nodes running Linux—can be converted to a hybrid configuration, where one half of the compute nodes run Oracle VM Server and serve as a virtualized data center, while the other nodes remain on Linux. While installing and configuring the Exalogic Elastic Cloud Software, you can choose the rack half (top or bottom) that should be virtual.

In addition, a hybrid machine can be converted to a full-virtual configuration.

The following hybrid configurations are supported:

  • Eighth: Top 2 or bottom 2 compute nodes

  • Quarter: Top 4 or bottom 4 nodes

  • Half: Top 8 or bottom 8 nodes

  • Full: Top 14 or bottom 16 nodes

    Note:

    "Bottom" in this context refers to the compute nodes starting from node 1. For example, in the case of a quarter rack, nodes 1 through 4 make up the bottom half, and nodes 5 through 8 the top half.

Conversion from a physical configuration to a hybrid configuration and from hybrid to full virtual configuration, requires reimaging compute nodes from Oracle Linux to Oracle VM Server. Oracle recommends strongly that any such conversion be performed by fully trained, qualified Oracle personnel or by formally accredited Oracle partners. For more information, contact Oracle Advanced Customer Support (http://www.oracle.com/acs).

1.1.3 Fewer VMs for the Exalogic Control Stack

The number of VMs that make up the Exalogic Control stack has been reduced from five in EECS 2.0.4 to three in 2.0.6.

The Exalogic Control stack in EECS 2.0.6 consists of the following VMs:

  • Two Proxy Controller VMs

  • Exalogic Control VM: This VM hosts the Enterprise Controller, Oracle VM Manager, and Oracle Virtual Assembly Builder components. The same VM also hosts the Oracle database used by EM Ops Center and Oracle VM Manager. In earlier releases, each of these components was hosted on a dedicated VM.

Note that, in EECS 2.0.6, the Exalogic Control VMs are distributed over fewer compute nodes: two nodes in the case of a fresh EECS 2.0.6 installation, and three nodes if you upgrade from EECS 2.0.4 to 2.0.6.

1.1.4 LVM-Based Disk Partitioning on Guest vServers

The local disks on guest vServers that you create by using the EECS 2.0.6 Guest Base Template use Logical Volume Manager (LVM)-based partitioning, which gives the administrator more flexibility in allocating disk space to applications and users.

The appendix titled "Managing LVM Partitions on Guest vServers" in the Oracle Exalogic Elastic Cloud Administrator's Guide provides a few examples for LVM-based partition management.

For more information about using LVM, go to http://www.howtoforge.com/linux_lvm.

1.1.5 Exabus/IMB 1.1 and Exalogic Java 1.1.6

EECS 2.0.6 includes version 1.1 of Oracle Exabus/IMB, which provides improved performance and stability for Java message passing in Oracle Fusion Middleware running on Exalogic. In addition, version 1.1 addresses bugs in the earlier version. EECS 2.0.6 also includes version 1.1.6 of Oracle Java loader for Exalogic.

1.1.6 Configuring DNS for EoIB Networks

While creating an EoIB network, you can specify DNS settings that should be used to access guest vServers (associated with that network) from outside the Exalogic machine by using host names.

For more information, see the "Creating External EoIB Networks" section in the Oracle Exalogic Elastic Cloud Administrator's Guide.

1.1.7 OSWatcher Included in the Base Image and Guest Template

A collection of shell scripts (called OSWatcher) is included in the Exalogic Guest Base Template and the Exalogic Base Image. OSWatcher collects and archives operating system and network metrics at fixed intervals, by running UNIX utilities such as vmstat and netstat. This data can be useful in diagnosing performance issues.

By default, OSWatcher is installed, configured, and running on all the Oracle VM Server nodes on an Exalogic machine and on all guest vServers created using the EECS 2.0.6 Guest Base Template. The utilities that OSWatcher invokes run as background processes and impose minimal incremental overhead on the system. By default, OSWatcher invokes the data-collection utilities once every 30 seconds. It archives the data for the last 48 hours on the local disk in the /opt/oswbb directory. You can configure the data-collection interval, the archive-retention period, and the archive location. You can also disable OSWatcher.

For more information about configuring, disabling, and enabling OSWatcher, see the OS Watcher Black Box User's Guide in the My Oracle Support document 1531223.1.

1.1.8 imagehistory and imageinfo Commands Included in the Base Template

The Exalogic Guest Base Template in EECS 2.0.6 includes the following commands:

  • imageinfo: You can use this command to identify the image and build version of a guest vServer, as shown in the following example:

    # imageinfo
    Exalogic 2.0.6.0.0 (build:r227074)
     
    Image version       : 2.0.6.0.0
    Image build version : 227074
    Kernel version      : 2.6.32-400.26.3.el5uek
    
  • imagehistory: You can use this command to view the lifecycle history of a guest vServer, as shown in the following example:

    # imagehistory
     
    Image version       : 2.0.6.0.0
    Image build version : 227074
    Image mode          : fresh
    

    Note that in the case of a vServer for which the template was patched or upgraded, Image mode will show update and the kernel version will be displayed for all the image versions. In addition, for every upgrade and patch applied, the date and time when the image was activated and the status will be displayed, as shown in the following example:

    # imagehistory
    Image version       : 2.0.4.0.0
    Image build version : 221802
    Kernel version      : 2.6.32-200.21.2.el5uek
    
    Image version       : 2.0.6.0.0
    Image build version : 227074
    Kernel version      : 2.6.32-400.26.3.el5uek
    Image activated     : 2013-08-28 22:52:45 +0300
    Image mode          : update
    Image status        : SUCCESS
    

1.1.9 Linux Server-Based Exalogic Guest Base Template

The Exalogic Guest Base Template in EECS 2.0.6 contains a Linux server installation, as opposed to a Linux desktop installation in earlier releases. With this change, you can run Oracle Fusion Middleware, Oracle Database, and Oracle applications on guest vServers without installing additional modules that are necessary for Linux servers.

1.1.10 Detailed View of Resource-Allocation by Account

Cloud Admin users can get a consolidated view of the resources—vCPUs, storage, and memory—allocated to each account through the Accounts table in the Exalogic Control browser user interface.

For more information, see the "Examining the vDC" section in the Oracle Exalogic Elastic Cloud Administrator's Guide.

1.2 New Features and Enhancements in EECS 2.0.4

This section lists the new features and enhancements in EECS 2.0.4.

1.2.1 CPU Oversubscription

To improve hardware utilization and to facilitate denser consolidation of applications, the Exalogic vDC can be configured to support more vCPUs than the available CPU threads—that is, the CPU resources can be oversubscribed.

For more information, see the "Configuring CPU Oversubscription" section in the Oracle Exalogic Elastic Cloud Administrator's Guide.

1.2.2 High Availability for Guest vServers

When you create a vServer, high availability is enabled by default for the new vServer. If a HA-enabled vServer crashes, the failed vServer is restarted automatically on the same compute node.

For more information, see the "Managing High Availability of vServers" section in the Oracle Exalogic Elastic Cloud Administrator's Guide.

1.3 Known Issues

This section lists the known issues in the Exalogic Elastic Cloud Software. The issues listed in this section are applicable to all EECS releases unless noted otherwise.

1.3.1 Shared Storage Access for Customer vServers

vServers created by customers can access shared file systems in Exalogic using the IPoIB-vserver-shared-storage network. Multiple accounts in the Exalogic virtualized environment use shared binaries across vServers. Oracle recommends that customers set access guidelines and restrictions accordingly.

1.3.2 Usage of the Maintenance Mode Feature

In the Exalogic Control BUI on an Exalogic machine running EECS 2.0.4, the Place in Maintenance Mode action is enabled and clickable for any compute node that you select. However, you must not use that action for compute nodes on which any vServer is running.

In EECS 2.0.6, the Place in Maintenance Mode action is disabled when you select a compute node on which any vServer is running.

1.3.3 Removing a Public Network from an Account Is Not Supported

Affected release: 2.0.4

Exalogic does not support removing a public network from an account.

Workaround

You can set the network resource limit for the account to zero, as follows:

Note:

Before setting the network resource limit for an account to zero, ensure that all the vIPs of the network have been deallocated as described in the "Deallocating Virtual IPs for an Account" section of the Exalogic Elastic Cloud Administrator's Guide.

  1. Log in to the Exalogic Control BUI as a user with the Cloud Admin role.

  2. Select vDC Management from the navigation pane on the left.

  3. Expand Accounts and select the account from which you want to remove the network.

    The dashboard for the account is displayed

  4. From the Actions pane on the right, select Update Account.

    The Update Account wizard is displayed.

  5. Click Next.

  6. Scroll down to the Public Networks Resource Limits table.

  7. Select the Limit field for the network you want to remove and set it to 0.

  8. Click Next.

  9. Click Finish.

1.3.4 Suspended State Is Not Supported for vServers

Exalogic does not support the suspended state for vServers.

Workaround

If a vServer is in the suspended state, you can do the following to return it to the running state:

  1. Log in to the Oracle VM Manager console as admin:

    https://host:7002/ovm/console/
    

    host is the host name or IP address of the vServer hosting the Oracle VM Manager component of Exalogic Control.

  2. For EECS 2.0.4 (or earlier releases), do the following:

    1. Under the Home view, select the vServer that is in the suspended state in the Server Pools folder.

    2. Right-click on the vServer, and select Kill from the context menu.

    3. Right-click on the vServer, and select Start from the context menu.

    For EECS 2.0.6, do the following:

    1. In the toolbar above the navigation pane on the left, click Find (magnifying glass icon).

      The Find dialog box is displayed.

    2. In the Type field, select Virtual Machines.

    3. In the Result field, select the vServer that you want to kill, and then click OK.

    4. On the resulting page, right-click on the required vServer, and select Kill from the context menu.

    5. Right-click on the vServer, and select Start from the context menu.

  3. Log out of the Oracle VM Manager console.

1.3.5 Full Job Log Not Visible in the Exalogic Control BUI

Affected release: 2.0.4

For a job initiated by a user with the Cloud User role, the Exalogic Control BUI does not display the full job log when the same Cloud User or the Exalogic System Admin user is logged in.

Workaround

To view the full job log, log in as the root user.

1.3.6 Cloud Users Cannot Act on Alerts

Affected release: 2.0.4

Users with the Cloud User role cannot act on alerts displayed in the Incidents tab of the Exalogic Control BUI.

Workaround

To act on alerts displayed in the Incidents tab of the Exalogic Control BUI, log in as a user with the Cloud Admin role.

1.3.7 Exalogic Control BUI Should Not Be Used for Upgrading Firmware

Affected release: 2.0.4

In the Assets accordion of the Exalogic Control BUI, when a compute node, switch, or PDU is selected, the Update Firmware action is available in the Actions pane. However, do not use the Update Firmware action in the Exalogic Control BUI to upgrade the firmware of Exalogic assets.

To apply firmware updates provided in Exalogic Patch Set Updates (PSUs) and patches, follow the instructions in the README that is included in the relevant PSU or patch.

1.3.8 Limit on Number of Volumes Attached to a vServer

When a guest vServer is created, a storage volume is created and associated with the vServer by default. This volume contains the /, root, and swap disks. In addition to this storage volume, a maximum of three more volumes can be attached to a single vServer. Attaching more than three additional volumes to a vServer is not permitted, and attempts to do so will fail with an appropriate error message.

1.3.9 Cannot Add Alert Monitoring Rules for vServers Through the vDC Management Accordion

Affected release: 2.0.4

In the Exalogic Control BUI, alert-monitoring rules for vServers cannot be added through the vDC Management accordion—that is, by selecting a vServer, selecting the Monitoring tab, and clicking the Add Alert Monitoring Rule button (plus icon).

Workaround

You can add alert-monitoring rules for vServers by selecting the vServers in the Assets accordion, selecting the Monitoring tab, and then clicking the Add Alert Monitoring Rule button (plus icon) there.

1.3.10 Time Zone of Guest vServers is Hard-Coded to PST

When a vServer is created, the time zone is hard-coded to Pacific Standard Time (PST).

Workaround:

Do one of the following:

  • Set the required time zone in the base template by modifying it.

  • Set the required time zone in the vServers.

1.3.11 Exalogic Control BUI May Select the Wrong vServer for Deletion

Affected release: 2.0.4

In isolated cases, when a user selects a vServer in the Exalogic Control BUI and deletes it, Exalogic Control may select a different vServer. This issue occurs due to BUI refresh issues.

Workaround

While deleting a vServer, before clicking the final Delete button, carefully verify the name of the vServer that is displayed as selected for deletion in the Confirmation screen.

1.3.12 "Navigate to Selected Oracle VM Server and Virtual Server" Buttons Do Not Work

When you log in to the Exalogic Control BUI as a user with the Cloud Admin or Exalogic Systems Admin role, if you expand Administration in the navigation pane on the left, you can select the Oracle VM Manager and view a list of Oracle VM servers and virtual machines on the Summary tab. However, in the Oracle VM Servers and Virtual Machines sections of the Summary tab, the Navigate to the selected Oracle VM Server and Navigate to the selected Virtual Server buttons—with the Icon with the text GOTO icon—do not work.

Workaround

You can navigate to a specific vServer by doing the following:

  1. In the navigation pane on the left, select Assets.

  2. In the search field, enter the name of the vServer you want to find.

  3. Click the Down arrow button adjacent to the search field.

1.3.13 Uploading a Server Template May Fail

Uploading server templates may fail in the following circumstances:

  • You are uploading a server template larger than the amount of free space on the Exalogic Control vServer.

  • Multiple users are uploading server templates simultaneously with a total size greater than the free space on the Exalogic Control vServer.

1.3.14 Late Lock Exception

Affected release: 2.0.6

When multiple vDC operations are running concurrently—such as vServer deletion and creation jobs running in parallel or OVAB assembly operations running in parallel, at times, some of the operations may fail.

For example, when five users concurrently register OVAB assemblies, while some of the assembly registration operations are completed successfully, a few may fail with an error message in the Exalogic Control console, as shown in the following example:

Result : Import Ovf Assembly from URL:
http://192.168.20.13:8003/tmpResourceFile.1370426698682.ova Failed. Exception occurred: cannot create template from assembly: OVMAPI_A019E Server operationsucceeded but job failed due to a late lock exception. 

This error is more likely to occur for operations involving vServers with a single vCPU than for vServers with more than one vCPU.

If this issue occurs, try the failed job again.

1.3.15 vServer Shutdown Failure

Affected release: 2.0.6

At times, a vServer shutdown job may hang and remain in a suspended mode until Exalogic Control times out the operation. Until the operation is timed out, other jobs in the queue may be blocked.

Workaround

Kill the vServer by using Oracle VM Manager, as follows:

  1. Log in to the Oracle VM Manager BUI as admin:

    https://elcontrol_vm:7002/ovm/console/
    

    elcontrol_vm is the host name or IP address of the Exalogic Control vServer.

  2. Select the Servers and VMs tab.

  3. In the toolbar above the navigation pane on the left, click Find (magnifying glass icon).

    The Find dialog box is displayed.

  4. In the Type field, select Virtual Machines.

  5. In the Result field, select the vServer that you want to kill, and then click OK.

  6. On the resulting page, right-click on the required vServer, and select Kill from the context menu.

  7. Log out of the Oracle VM Manager console.