test

Sun Open Telecommunications Platform 1.1 Release Notes

Chapter 2 Release Notes

This chapter describes the Open Telecommunications Platform (OTP) issues that are known to be problems. The following topics are discussed:

Open Telecommunications Platform 1.1 Limitations

The following are known limitations of Open Telecommunications Platform 1.1:

Installation and Configuration Issues

This section describes known Open Telecommunications Platform installation and configuration issues.

Deletion of the OTP Plug-in Requires the Deletion of All Hosts (6458703)

Before you can delete the OTP plug-in from an external OTP installation server or a production OTP host, you must first delete all hosts for which you ran plans.

  1. Log in to the server from which you ran the Plans to add additional OTP hosts.

  2. Delete all of the target hosts only for each host for which you ran Plans from the OTP plug-in.

  3. Delete the OTP plug-in.

Resource Group otp_system_group and Resource N1SPSMS-HA Fails Under Heavy System Load (6507395)

During load test on memory, the resource group otp_system_group fails with the error message Error--stop failed, and the resource N1SPSMS-HA fails with the error message Stop failed Faulted.

Workaround:

  1. Log in as root on the clustered OTP host on which the OTP system management service is installed and running.

  2. Restart the remote agent by typing /etc/init.d/n1spsagent restart. If the remote agent is not restarted, then the service provisioning service on the first OTP host will not work properly.

  3. Type /usr/cluster/bin/scrgadm -c -g otp-system-rg -y RG_system=false to set the system property for the otp-system-rg resource group to false.

  4. Type /usr/cluster/bin/scswitch -F -g otp-system-rg to take the remote group offline.

  5. Type the following commands in the sequence shown to disable cluster resources.

    1. /usr/cluster/bin/scswitch -n -j otp-spsms-rs

    2. /usr/cluster/bin/scswitch -n -j otp-spsra-rs

    3. /usr/cluster/bin/scswitch -n -j otp-sps-hastorage-plus

    4. /usr/cluster/bin/scswitch -n -j otp-lhn

  6. Type /usr/cluster/bin/scswitch -u -g otp-system-rg to put the remote group into the unmanaged state.

  7. Type /usr/cluster/bin/scrgadm -c -j otp-spsra-rs -x Stop_signal="15" to change the Stop_signal property of the remote agent resource to 15.

  8. Type /usr/cluster/bin/scrgadm -c -j otp-spsms-rs -x Stop_signal="15" to change the Stop_signal property of the management service resource to 15.

  9. Type /usr/cluster/bin/scswitch -o -g otp-system-rg to put the remote group into the managed state.

  10. Type /usr/cluster/bin/scswitch -Z -g otp-system-rg to bring the remote group back online.

  11. Type /usr/cluster/bin/scrgadm -c -g otp-system-rg -y RG_system=true to set the system property for the otp-system-rg resource group to true.

OTP Does Not Validate All User Input During Installation (6515467)

When you enter invalid inputs during OTP installation, installation might succeed. However, OTP services might not be properly configured.

Workaround: Make certain that you provide valid values when entering OTP plan information. The Sun Open Telecommunications Platform 1.1 Installation and Administration Guide provides OTP Plan worksheets for you to record installation information for each host prior to the installation process, which you should then use during installation to ensure the data you provide is correct. Print a worksheet for each host and then record and verify the information for each host prior to installing OTP.

Installation of OTP Fails If IPMP Is Not Correctly Configured (6549401)

OTP installation fails at the “OTP System Management and Provisioning Service” step.

There are two types of IPMP configurations: Link-based failure detection and Probe-based failure detection.

OTP supports only probe-based failure detection. The installation fails if IPMP is configured in the link-based failure detection mode.

Workaround: Before installing OTP, make sure that IPMP is configured in the probe-based failure detection mode. If you want to configure IPMP in the link-based failure detection mode, you can do it after OTP installation.

OTP Installation From External Server Through GUI Displays an Error Message (6536993)

Service Management Facility (SMF) is registered with the old calendar service rpc-100083. After applying the patch, SMF gets registered with the new calendar service. This registration happens only after the first reboot. Due to this registration, when you install OTP from external server through GUI, you might see an error message as follows:

Failed to update state of instance svc:/network/rpc-100083_1/rpc_tcp:default in repository: entity not found

This error message appears only once during the first reboot.

Warning Messages on Console During OSP Installation (6529175)

During OSP installation, you might see some warning messages on the console. These messages do not affect the functionality of the OSP plug-in and can be ignored.


Solaris: Configuring JumpStart boot for target node
                 Starting SMF services for JumpStart
Solaris: Configure DHCP build
                 Adding install client
WARNING:          JS_DHCP_VENDOR is not defined in jumpstart.conf
WARNING:          This value a space separated list of the uname -i
WARNING:          outputs of the machines you want to install.
WARNING:          Supporting 'SUNW.i86pc' by default.

OS Installation Fails After the Initial Reboot (6549371)

Assume that you have done a initial setup as follows:

If you reboot the host after this initial setup, OSP and Jet directories become unavailable. Hence, OS installation fails after the initial reboot.

Workaround: Run the following command after the initial reboot.

# /usr/sbin/svcadm enable network/nfs/server

Integration Issues

This section describes known Open Telecommunications Platform integration issues.

OSP Deployment Plan Completes Before OS Is Installed (6530773)

When the OSP plan is run to install the OS, the plan completes before the OS is installed. Hence, the OS installation will still be in progress after the OSP plan completes.

Workaround: Monitor the progress of OS installation using the following steps.

  1. In the Common Tasks section of the GUI, select OS Provisioning.

  2. On the OS Provisioning Common Tasks page, click Status in the OS Provisioning Administration Tasks section.

  3. On the Plans Details page, click Run.

  4. On the Plan Details Run page, select the target host on which you provisioned the OS.

  5. Click Run Plan (includes preflight).

  6. Follow the Details links to view the status.

OSP Plug-in Cannot Import Two OS Images Simultaneously (6531142)

The OSP plug-in does not import two OS images simultaneously. If you try to import two OS images simultaneously, the following message appears:

Connect and lock hosts

Workaround: Wait for the first OS image to be imported before importing the second OS image.

OSP Plug-in Does Not Take a Shared File System as OS Image Path (6531133)

The OSP plug-in is not highly available. Hence, OSP plans fail if a shared file system is given as OS image path. Due to this issue, you cannot place OS images on the shared disks.

Error Message:


osp_create_solaris_image.sh: 'targetnode/export/home/Solaris10U2' 
'/var/otp/osp_image/Solaris10_u2_sparc' '10' 'Solaris10_u2_sparc' ' /opt/SUNWjet'
Created loopback device /dev/lofi/1 for 
target node/export/home/Solaris10U2/sol-10-u2-ga-sparc-dvd.iso
Failure in mounting /var/otp/osp_image/3047/slices/s0
Filesystem  does not appear to be mounted.
removing device /dev/lofi/1
removing directory /var/otp/osp_image/3047
osp_create_solaris_image.sh: ERROR -- copy_solaris_media failed

cfgadm Does Not Warn if Metaset Is Used (6536514)

The cfgadm command provides configuration administration operations on dynamically reconfigurable hardware resources. The cfgadm command on the OTP cluster node does not warn if the metaset is being used and unconfigures the controller, which results in unstable OTP Services.

Hosts Cannot Be Deleted From the Hosts List of External Server (6527578)

When you try to delete hosts from the hosts list of external server through GUI, you might get the following error message:


Unable to delete the host
At least one delete operation failed

Workaround: Restart the master server on the specific node and remove the host.

OSP ALOM Create Plan Does Not Validate the Password (6537015)

OSP ALOM create plan does not verify whether the admin password entered is correct or not. The plan succeeds with any password.

OSP Uses the IP Address of Installation Server as the Default Router (6539435)

If you do not explicitly specify the IP address of the default router in the Provision OS profile, OSP will use the IP address of installation server to create a /etc/defaultrouter file.

Workaround: Specify the IP address of the default router in the Provision OS Profile or modify the /etc/defaultrouter file to make it point to the correct router information and reboot the system.

EIS compliance requires the /etc/defaultrouter file to be present.

Messages On Console When Host Reboots (6535460)

Runtime error messages are displayed on the console when the host reboots.

Error snippet:


libpkcs11: Unable to contact kcfd: Bad file number
libpkcs11: /usr/lib/security/pkcs11_softtoken.so unexpected failure 
in ELF signature verification. System may have been tampered with. 
Cannot continue parsing /etc/crypto/pkcs11.conf
com.sun.hss.domain.PersistentMBeanServer.loadMBeans : 
Error loading MBeans from persistent store
com.sun.cacao.element.ElementSupport.setAdministrativeState : 
caught throwable when starting : RuntimeException thrown in postRegister method

These messages are completely harmless and can be safely ignored.