SOFTPAQ NUMBER: SP14375 PART NUMBER: N/A FILE NAME: N/A TITLE: Compaq Insight Manager Agents and Tools - SCO UnixWare 7 NonStop(TM) Clusters VERSION: 4.90a LANGUAGE: English CATEGORY: Software Solutions DIVISIONS: Systems PRODUCTS AFFECTED: Compaq Servers running SCO UnixWare 7 OPERATING SYSTEM: SCO UnixWare 7.1.1 SYSTEM CONFIGURATION: N/A PREREQUISITES: - Compaq Enhanced Feature Supplement (EFS) for SCO UnixWare 7, Version 7.38 The nsc package - Version 7.1.1 or 7.1.1b EFFECTIVE DATE: September 1, 2000 ELECTRONIC DISTRIBUTION ALLOWED: Yes SOFTPAQ UTILITY VERSION: 4.1 SUPERSEDES: nscups, nscvu, cmafdtn, cmastor, cmanic, cmasvr versions 4.80. DESCRIPTION: Creates diskettes containing Compaq Insight Manager Agents and Tools - SCO UnixWare 7 NonStop(TM) Clusters, Version 4.90a. IMPORTANT: This softpaq can be installed on the 7.1.1 or later versions of SCO UnixWare 7 NonStop Clusters. It upgrades the version 4.80 CIM agents and tools or installs an initial version of CIM Agents and Tools for SCO UnixWare 7 NonStop Clusters version 7.1.1. HOW TO USE: Perform steps 1-5 under DOS, Microsoft Windows 3.1, Windows95 or Windows NT. Perform steps 6-9 under SCO UnixWare 7 NonStop Clusters. 1. Obtain 4 IBM-formatted 1.44-MB diskettes. 2. Download the SoftPaq to a clean directory on your hard drive and change to that directory. The downloaded file is an executable with a filename based on the SoftPaq Number above. 3. From that drive and directory, execute the downloaded file and follow the on-screen instructions. Seven files will be created: SP14375.TXT, SP14375._01, SP14375._02, SP14375._03, SP14375._04, sp14375.cva, and QRST5.EXE. 4. Execute QRST5.EXE and at the prompt, insert the formatted diskettes as prompted in the diskette drive to create the four diskettes containing Compaq Insight Manager Agents - SCO UnixWare 7 NonStop Clusters, Version 4.90a. Label the diskettes with their numbered sequence. The first diskette contains the following: * nsc1013, which is required for the nscccm package. * nscvu, the NonStop Clusters Verification Utility. * nscups, the NonStop Clusters Uninterruptible Power Supply utility. * nscccm, the Common Cluster MIB, which must be installed. The second diskette contains cmafdtn, the Foundation Agents package. The third diskette contains cmastor, the Storage Agents package. The fourth diskette contains the following packages: * cmasvr, the Server Agents package. * cmanic, the Networking Agents package. 5. After the diskettes have been created, you can delete the SoftPaq file downloaded in step 2 and the files unpacked in step 3. 6. To install the Compaq Insight Manager Agents - SCO UnixWare 7 NonStop Clusters, log into your cluster as root. 7. Ensure that the EFS drivers are version 7.38 or higher by issuing one of the following commands : For Compaq Health Driver Update, enter # pkginfo -l ptfcpqw 8. Insert the first newly created diskette into the cluster's default diskette drive and enter: pkgadd -d diskette1 Select all packages, and follow the on-screen directions. Ignore any instructions concerning the download on nsc1013. 9. Insert the second floppy diskette, and enter: pkgadd -d diskette1 Select all packages, and follow the on-screen directions. 10. Insert the third floppy diskette, and enter: pkgadd -d diskette1 cmastor Follow the on-screen directions. 11. Insert the last floppy diskette, and enter: pkgadd -d diskette1 cmasvr cmanic Follow the on-screen instructions. 12. Remove the last of the four diskettes. ----------Notes------------------------------------------------- ASR Notes: Do not install the EFS Compaq Advanced Server Recovery ptfcpqasr package. Uninstall the ptfcpqasr package if it has been previously installed. Ensure that the Advanced Server Recovery Option has been disabled. to disable ASR, press F10 when prompted during the boot and select System Configuration-->Configure Hardware-->Review or Modify Hardware Settings-->Step 3: View or Edit Details. Scroll to the Embedded - Automatic Server Recovery (ASR) section and set the Automatic Server Recovery option to disabled. Follow the on-screen instructions to save and exit then reboot the system. NSCVU Notes: Refer to the nscvu(1M) manual page for usage information. The nscvu command: Outputs the incorrect value of cache for ML330 systems. Can take as long as 30 minutes to display system information on systems with a large number of mounts. Can show multiple identical mount points on fiber-connected storage systems that use a large number of mounts. Is not supported on RSO configurations. Can provide incorrect results if a root node is down. Can take a long time to execute or start failover testing on a system with a large number of mounts. Power Management Notes: Configure UPS-initiated shutdown by modifying the os_shutdown, ups_serial_port, and ups_log_file parameters in the /opt/compaq/etc/nscupsd.cfg configuration file. os_shutdown Specifies the battery backup power remaining when a cluster-wide shutdown is initiated. For example, an entry of os_shutdown=15 indicates that a cluster-wide shutdown is initiated when the UPS has 15 minutes of battery backup power remaining. Measure the time required for a clean shutdown of the cluster under peak operating conditions to ensure that the shutdown time is adequate. ups_log_file Specifies the file that contains event information related to UPS state transitions. The default parameter is /var/spool/compaq/nscupsd.log and should not be modified. ups_serial_port Specifies a list of ports that are separated by colons and semicolons. This list specifies the serial ports to which the UPSs are connected and the combination of UPS signals for shutting down the cluster. Colon-separated serial ports create a pair of UPSs in which both of the UPSs must signal that they are low on power before a cluster shuts down. Such a group of UPSs are referred to as logical UPSs. A logical UPS identifies UPSs that are grouped together to provide redundant power to a cluster. The drain of a single UPS within a logical UPS should not result in the loss of any key cluster resources. Semicolon-separated serial ports identify a list of UPSs or logical UPSs. A low-power indication by any of those UPSs results in a cluster-wide shutdown. This parameter is used when a cluster spans multiple power domains and the loss of any one domain results in a cluster-wide shutdown to protect the cluster. NOTE: A serial connection to a UPS is used in determining shutdown if the node with the serial port is an active member in the cluster. Example ups_serial_port Configurations For a two-node cluster or single-rack cluster with two UPSs, the cluster should shut down when both UPSs are low on power. To have both UPSs signal that they are low before shutting down the cluster, combine both UPSs into a single logical UPS. This results in a ups_serial_port configuration of: ups_serial_port=/dev/tty00.1:/dev/tty00.2 Where /dev/tty00.1 is the device identifier for the node 1 serial port tied to UPS1 and /dev/tty00.2 is the device identifier for the node 2 serial port tied to UPS2. In a multiple-rack cluster that uses fully redundant power in all key cluster components, the cluster should shut down only if both UPSs in a single rack lose power. For both UPSs within a rack to signal that they are low before shutting down the cluster, both UPSs within a rack are joined into a single logical UPS. This results in an ups_serial_port configuration of: ups_serial_port=A:B;C:D; In which case the following is true: A is the device identifier for the node serial port tied to UPS1A (/dev/tty00.1 for port 0 on node 1) B is by the device identifier for the node serial port tied to UPS1B (/dev/tty00.2 for port 0 on node 2) C is the device identifier for the node serial port tied to UPS2A (/dev/tty01.1 for port 1 on node 1) D is by the device identifier for the node serial port tied to UPS2B (/dev/tty01.2 for port 1 on node 2) Starting the UPS-Initiated Shutdown Software After UPS-initiated shutdown has been configured as described in the preceding sections, restart the UPS daemon by entering the following two commands: /etc/init.d/nscups stop /etc/init.d/nscups start CIM Agent Notes: * Agent startup can take up to 15 minutes when a node is rebooted. * Agent startup can take up to 10 minutes per node at CIM installation. * Use the following command syntax to perform the agent start or stop operation in the background: /etc/init.d/cpqsnmp start | stop * Use the following command syntax to perform the agent start or stop operation on a specific node: /etc/init.d/cpqsnmp start_for | stop_for * Use the following command syntax to operate on a specific node: /etc/init.d/cpqsnmp start_for/stop_for * Occasionally, not all daemons on all nodes will start properly upon initial installation. If this situation occurs, either reboot the cluster, or enter the following commands: 1. Enter: onall /etc/init.d/cpqsnmp stop_for 2. Wait five minutes, then enter onall /etc/init.d/snmp_node stop 3. Wait one hour, then enter: onall /etc/init.d/cpqsnmp start_for A known problem exists where heavy web use may cause the web daemon to consume excessive resources. If this should occur, stop the daemon with the following command: onall /etc/init.d/cmaweb stop. You can prevent the daemon from starting at boot time with the following command mv /etc/init.d/cmaweb /etc/init.d/cmaweb.disabled. * The cache size is incorrectly displayed for ML330 systems. * In a terminal window on a graphical console, the pkgadd cmafdtn command, which relies on having /sbin in the PATH variable will not work properly. To correct this, set the PATH environmental variable before executing the pkgadd command. The following is a Bourne Shell example: # PATH=$PATH:/sbin # export PATH * CIM may not show all of the health information for node 2 of a IP interconnect system with NSC 7.1.1+IP without SCO PTF nsc1011. ------------------------------------------------------------------ Copyright 2000 Compaq Computer Corporation. Product names mentioned herein may be trademarks and/or registered trademarks of their respective companies.