Sun xVM Ops Center 1.1.1 Release Notes |
|
Contents |
Revision History:
September 16, 2008 - Added BIOS issue with the M2 Sun Netra™ x4450 servers in General Issues
July, 2008 - General Availability
New Features
New features for Sun xVM Ops Center 1.1.1 include:
- Upgrade capabilities
- Improved quality
Installing Sun xVM Ops Center 1.1.1
- Get the software
- Follow the installation steps to install Sun xVM Ops Center software. Perform the installation and upgrade procedures described in these pages.
If You Have Sun xVM Ops Center 1.1 Installed
If you have Sun xVM Ops Center 1.1 installed, you can use the upgrade feature to upgrade to version 1.1.1. Follow the upgrade procedures to perform the upgrade.
If You Have Sun xVM Ops Center 1.0, 1.0.1, or 1.0.2 Installed
- If you do have Sun xVM Ops Center 1.0, 1.0.1, or 1.0.2 installed on your system, use the procedures described on the Special Procedure for Early Access Installations page to uninstall Sun xVM Ops Center software before you proceed with the version 1.1 installation.
- Install the 1.1 version of the software, using the Sun xVM Ops Center 1.1 installation procedures.
- Upgrade the Sun xVM Ops Center Server to version 1.1.1 using the upgrade procedures.
Fixed Issues
The following issues are fixed:
| Bug ID | Description |
|---|---|
| 6684801 | Agent version is not displayed on the Update Gear agentry page |
| 6687509 | Problem encountered when restoring the Satellite |
| 6702949 | System Catalog is eporting false package version |
| 6705136 | Upgrade from 1.0.2 to 1.1 does not work properly |
| 6705525 | Uninstall did not remove all files |
| 6707477 | sw_mgmt_serv Missing channels.xml causes problems with major update channel functionality |
| 6713069 | Click through license file for 1.1.1 release updated |
| 6720472 | Update bundle created for 1.1.1 |
| 6722588 | Vtreee did not get updated with the "url value" after the upgrade |
| 6722509 | The timeout for satadm start/stop commands needs to be increased |
| 6723293 | xVM Ops Center uninstalls 100s of packages when rolling back install of a few packages |
| 6727089 | The local components disappear after the Upgrade from 1.1 to 1.1.1 |
Known Issues
The following known issues and bugs affect the operation of the Sun xVM Ops Center 1.1.1 release. Bugs and issues are grouped into the following categories:
- Installation
- General
- Gear Module - Discovery and Registration
- Updates Module
- Provisioning Module
- Management Module
Installation
The following are known installation issues:
| Bug ID | Description | Workaround |
|---|---|---|
| 6633596 | Install might hang at "10) Install Core Channel components" step | Reboot the system so that the SMF starter svc.startd is restarted. Then retry the install to continue where it left off. |
| 6664928 | When the satellite and proxy software is installed on operating environments that do not meet the requirements, the installation does not clearly state the missing packages. | A partial list of required packages is SUNWgcc, SUNWgccruntime, SUNW,apchr, SUNWapch2s, SUNWapch2d, SUNWapch2u, SUNWant, SUNWgetr, SUNWgetu, and SUNWperl-xml-parser. |
| 6702415 | On the control proxy page, the Satellite Registration does not start the proxy and close the wizard | Click the cancel button in the wizard, or logout (https://sat:9443/scportal/logout) and login again. Verify that the proxy starts. If the proxy does not start, use the following command to start the proxy from the command line: proxyadm start -w. |
| 6689967 | /var/opt/sun is temporarily 700/scndb/scndb during install, can cause install errors if install fails on Install Agents components step | If the installer fails on the Install Agent components step, the workaround is as follows: 1) Continue with uninstall. 2) before reinstalling, rm -rf /var/opt/sun 3) reinstall |
| 6702078 | After installation failure, uninstall then reinstall fails on linux | 1) Continue with uninstall. 2) before reinstalling, rm -rf /var/opt/sun 3) reinstall |
General
The following are known issues:
| Bug ID | Description | Workaround |
|---|---|---|
| M2 Sun Netra x4450 servers with BIOS below version S93_3B13 might have the same UUID. If two systems have the same UUID, Sun xVM Ops Center can only manage the first system. | Upgrade all x4450M2 servers to S93_B13 or higher before trying to discover them from Sun xVM Ops Center. | |
| 6656276 | proxyadm stop -w times out because all proxy processes have not exited. | Run svcs -pv svc:/application/management/common-agent-container-1:scn-proxyto see the processes associated with the proxy. If there is no java process, but a dsvclockd or an rsshcmd process is shown, then kill the process and the proxy will be completely shut down. |
| 6658616 | xVM Ops Center management operations on the satellite server and operating system cause problems. | Do not use xVM Ops Center to perform management operations on the server that runs the xVM Ops Center satellite. Do not enable update channel for the operating system hosting the xVM Ops Center satellite. |
| 6664997 | Local/embedded proxy controller does not properly start upon Satellite server boot | |
| 6662780 | xVM Ops Center update and provisioning operations on the satellite server and operating system cause problems. | Do not use xVM Ops Center to provision or update the firmware or operating system on which the satellite server is installed. Do not enable provisioning or updating for that gear. |
| 6644911 | Select uninstall action does not check the required selection all the times | |
| 6692408 | Login page sometimes requires you to log in again. This might occur if there is a long delay between visiting the login page and attempting to login. | Log in a second time. |
| 6670059 | n1sh will exit when user enters the escape keys in 'connect node <node_id>' | Invoke the 'n1sh connect node <node_ip>' from the command line rather than in the interactive mode. |
| 6657375 | BUI logout sometimes occurs even when user is working | |
| 6706792 | Linux - if database fails to start and user logs in, will be forced to re-register | Determine DB issue, correct, stop/start Satellite. Log back in and go through Satellite reg wizard again, then you will be brought to normal dashboard page |
| 6729083 | xVM Ops Center fails to create jobs if the host or job name is too long | Host and job name entries are limited to 32 characters. To avoid this issue, use host names that are less than 24 characters in length. |
| 6728676 | Unable to run jobs after installation | Refer to 6729083. Host and job name entries are limited to 32 characters. To avoid this issue, use host names that are less than 24 characters in length. |
Gear Module - Discovery and Registration
The following are known issues in the Gear module:
| Bug ID | Description | Workaround |
|---|---|---|
| 6613181 | Sometimes Wolf Blades (X6250) are not discoverable. This is due to the leftover PET settings on ELOM systems. | To remove the leftover PET settings, delete the server, or manually clean up the PET settings, before you uninstall Sun xVM Ops Center. |
| 6702274 | Discovery of invalid iprange of address should throw proper error message | Entering an IP range for discovery that is too large and includes more than the maximum number of IP addresses (16384) will result in an error indicating that the discovery job could not be submitted. The error does not indicate why the discovery job could not be submitted. The best way to avoid this is to enter smaller ranges for discovery. |
| 6701982 | A discovery criteria can be saved even when username/Paswd/IP/name field is empty. Saving a criteria without saving the password(s) leaves the empty password field in an uneditable state during the next user session. | Uncheck the specific protocol entry and then re-enter the protocol entry including the user credentials. |
Updates Module
The following are known issues in the Updates module:
| Bug ID | Description | Workaround |
|---|---|---|
| 6656896 | Suspension of Group jobs does not succeed for one usecase | The Suspend group job operation can be invoked when tasks other than the last task of an intermediate job of the group is in progress. This results in a correct suspension of the group job. |
| 6704390 | Channel id is incorrect" error message appears for some hosts I select for the "Create Catalog" wind | |
| 6723299 | System cloning/rollback using the Catalog Compare operatio may report incorrect task list. Attempting to clone or rollback a Solaris system with differing installed patches will not always perform the correct sequence of patch install/uninstall actions. As a result, the clone/rollback operation may either fail to initiate or result in an incorrect clone/rollback. | No known workaround |
| 6717595 | Update Jobs fail with download failure after temporary SOA failure | See below |
| 6726099 | Update agent core dumps on S10 U5 x86 agents | See below |
Workaround for 6717595
Diagnosis
- Check the Task Log for the failed target(s). It should report a failure like the following:
Download Phase - Download Failed. Check server logs for more details. Username/Password was marked as unauthorized during periodic check, download not attempted.
This may be caused due to temporary outages in the download site which XVMOC mistakenly interprets as an invalid SOA, even if the SOA credentials are correct.
Workaround
- Login to each agent and increase the retry interval by editing the /opt/SUNWuce/server/cgi-bin/.uce.rc and modifying the following property as shown:}
( all ) ( invisible.server.__general.knowledgebase_reconnect_interval, 600 );
- restart the agent
# svcadm disable -s update-agent # svcadm enable -s update-agent
- Re-run any jobs which have failed from the BUI
Note that this workaround will work only if your XVMOC update channel SOA credentials are actually valid. If they are invalid, the job will continue to fail. - All future jobs should now retry automatically
Workaround for 6726099
Diagnosis
The bug is due to corrupt patch information in the Staroffice packages
- run the patchadd command on the agent to check for the broken packages:
# patchadd -p|grep "Patch: 126411-01" Patch: 126411-01 Obsoletes: Requires: Incompatibles: tibles: Packages: SUNWstaroffice-core01 # patchadd -p|grep "Patch: 120186-13" Patch: 120186-13 Obsoletes: Requires: 6411-01 Incompatibles: Packages: SUNWstaroffice-core01 SU...
- You can see corrupted patchid "6411-01" or the broken value "tibles:
Workaround
Either remove the StarOffice packages or fix the broken pkginfo files.
- To fix the broken files, a script is available to correct the StarOffice packages for Solaris 10 Update 5 systems. Download the script here.
- Run the script to correct the files:
# ./fix_staroffice.sh Checking StarOffice packages for broken pkginfo lines Fixing broken line for patch 120186-13 in package SUNWstaroffice-base Fixing broken line for patch 120186-13 in package SUNWstaroffice-calc Fixing broken line for patch 120186-13 in package SUNWstaroffice-core01 Fixing broken line for patch 126411-01 in package SUNWstaroffice-core01 ... Fixing broken line for patch 120186-13 in package SUNWstaroffice-writer Fixing broken line for patch 120186-13 in package SUNWstaroffice-xsltfilter
Provisioning Module
The following are known issues with firmware and OS provisioning:
| Bug ID | Description | Workaround |
|---|---|---|
| 6660830 | OSP on multiple systems failing on one node in multi-proxy environment | This is a v20z hardware issue |
| 6698888 | ERROR: "/" does not fit on any disk | If you are using the default profile on a Linux proxy to install Solaris and if either Linux or Windows was installed previously, this error occurs because the rootdisk associated with the default profile is not being cleared on Linux. To resolve this issue, create a new profile associated with the specific disk. |
| 6653168 | Consecutive OSP jobs along with agent installation results in the second job being listed as partially successful | Undefined behavior can cause the agent to not work or not be seen correctly. The workaround is to unconfigure the agent manually. Discover the agent, then provision the OS. To unconfigure the agent, type the following on the system:
|
| 6701873 | Import SLES10 64bit iso images fail | Remove all files from the following directories before you begin importing images: /var/opt/sun/scn/images/os/iso/ /var/opt/sun/scn/images/os/flars/ |
| 6702686 | Importing an OS image fails , stating already exists in RHEL5 as Sat/Proxy. | |
| 6702063 | Create Firmware Image fails stating already exists in RHEL5 as Sat/Proxy |
Management Module
The following are known issues in the Management module:
| Bug ID | Description | Workaround |
|---|---|---|
| 6605211 | Reset server command powers off a X2100M2 management node | |
| 6658745 | Monitoring network failure of registered gears is not fully working |
Command Line Interface
With the exception of the following, CLI commands are not supported in this release:
Sign up or Log in to add a comment or watch this page.