NonStop NS-Series Operations Guide (H06.03+)
Table Of Contents
- What’s New in This Manual
- About This Guide
- 1 Introduction to Integrity NonStop NSSeries Operations
- When to Use This Section
- Understanding the Operational Environment
- What Are the Operator Tasks?
- Monitoring the System and Performing Recovery Operations
- Preparing for and Recovering from Power Failures
- Stopping and Powering Off theSystem
- Powering On and Starting the System
- Creating Startup and Shutdown Files
- Performing Preventive Maintenance
- Operating Disk Drives and Tape Drives
- Responding to Spooler Problems
- Updating Firmware
- Determining the Cause of a Problem: A Systematic Approach
- Logging On to an Integrity NonStop Server
- Service Procedures
- 2 Determining Your System Configuration
- 3 Overview of Monitoring and Recovery
- 4 Monitoring EMS Event Messages
- 5 Processes: Monitoring and Recovery
- 6 Communications Subsystems: Monitoring and Recovery
- 7 ServerNet Resources: Monitoring and Recovery
- 8 I/O Adapters and Modules: Monitoring and Recovery
- 9 Processors and Components: Monitoring and Recovery
- When to Use This Section
- Overview of the NonStop Blade Complex
- Monitoring and Maintaining Processors
- Identifying Processor Problems
- Recovery Operations for Processors
- Recovery Operations for a Processor Halt
- Halting One or More Processors
- Reloading a Single Processor on a Running Server
- Recovery Operations for a System Hang
- Enabling/Disabling Processor and System Freeze
- Freezing the System and Freeze-Enabled Processors
- Dumping a Processor to Disk
- Backing Up a Processor Dump to Tape
- Replacing Processor Memory
- Replacing the Processor Board and Processor Entity
- Submitting Information to Your Service Provider
- Related Reading
- 10 Disk Drives: Monitoring and Recovery
- 11 Tape Drives: Monitoring and Recovery
- 12 Printers and Terminals: Monitoring and Recovery
- 13 Applications: Monitoring and Recovery
- 14 Power Failures: Preparation and Recovery
- 15 Starting and Stopping the System
- When to Use This Section
- Powering On a System
- Starting a System
- Minimizing the Frequency of Planned Outages
- Stopping Application, Devices, and Processes
- Stopping the System
- Powering Off a System
- Troubleshooting and Recovery Operations
- Fans Are Not Turning
- System Does Not Appear to Be Powered On
- Green LED Is Not Lit After POSTs Finish
- Amber LED on a Component Remains Lit After the POST Finishes
- Components Fail When Testing the Power
- Recovering From a System Load Failure
- Getting a Corrupt System Configuration File Analyzed
- Recovering From a Reload Failure
- Exiting the OSM Low-Level Link
- Opening Startup Event Stream and Startup TACL Windows
- Related Reading
- 16 Creating Startup and Shutdown Files
- Automating System Startup and Shutdown
- Processes That Represent the System Console
- Example Command Files
- CIIN File
- Writing Efficient Startup and Shutdown Command Files
- How Process Persistence Affects Configuration and Startup
- Tips for Startup Files
- Startup File Examples
- Tips for Shutdown Files
- Shutdown File Examples
- 17 Preventive Maintenance
- A Operational Differences Between Systems Running GSeries and HSeries RVUs
- B Tools and Utilities for Operations
- When to Use This Appendix
- BACKCOPY
- BACKUP
- Disk Compression Program (DCOM)
- Disk Space Analysis Program (DSAP)
- EMSDIST
- Event Management Service Analyzer (EMSA)
- File Utility Program (FUP)
- Measure
- MEDIACOM
- NonStop NET/MASTER
- NSKCOM and the Kernel-Managed Swap Facility (KMSF)
- OSM Package
- PATHCOM
- PEEK
- RESTORE
- SPOOLCOM
- Subsystem Control Facility (SCF)
- HP Tandem Advanced Command Language (TACL)
- TMFCOM
- Web ViewPoint
- ViewPoint
- ViewSys
- C Related Reading
- D Converting Numbers
- Safety and Compliance
- Index
Disk Drives: Monitoring and Recovery
HP Integrity NonStop NS-Series Operations Guide—529869-001
10-13
Recovery Operations for Disk Drives
Unspared
defective
sectors
To check for unspared defective sectors with SCF:
-> INFO DISK $*, BAD, SEL started, sub magnetic
To check for unspared defective sectors with DSAP at a TACL prompt:
> DSAP $*
•
Recovery for DSAP is not needed.
•
Recovery for DCOM: use the SCF INFO DISK, BAD command on the
affected disk to obtain the bad sector address. Before restarting
DCOM, perform the CONTROL DISK, SPARE command.
For more information, see the Guardian Disk and Tape Utilities Manual.
Nearly full
database file
Recovery Operations for a Nearly Full Database File
on page 10-15
Performance
problems
Performance problems can have various causes, including path switches
or a cache size that is too small.
For information about disk load balancing and increasing cache size, see
the SCF Reference Manual for the Storage Subsystem.
Corrupt
$SYSTEM
disk
If both halves of your mirrored system volume become corrupted, use an
alternate system disk if one is available. For how to create an alternate
system disk, see the Integrity NonStop NS-Series Planning Guide.
•
For internal SCSI disk drives: if there is no alternate system disk and
you cannot load from the CONFBASE file, you might be able to
perform a tape load from a system image tape (SIT) to restore the
system image files to the $SYSTEM disk (SYSnn and CSSnn
subvolumes). Then load that image into processor 0 or 1.
A tape load reinitializes the disk directory. The disk directory is
overlaid with the directory from the tape. All files on that disk are
destroyed. Perform a tape load only with the advice of the Global
Customer Support Center or your service provider.
•
For M8xxx disk drives: you cannot perform a tape load from a SIT.
Failed disk
drives
•
Internal SCSI disks: the CSSI Web describes replacing disk drives.
•
M8xxx fibre-channel disks: these disks are FRUs and can be serviced
or replaced only by HP-trained service personnel.
Table 10-3. Common Recovery Operations for Disk Drives (page 2 of 2)
Problem Recovery