Tandem Failure Data System (TFDS) Manual

HP Tandem Failure Data System (TFDS) Manual—540122-003
1-1
1 Introduction to TFDS
First Failure Data Capture (FFDC) is an industry wide concept for capturing information
at the first instance of a particular failure, and avoiding redundant failure data collection
and problem-solving efforts. Tandem Failure Data System (TFDS) is the
implementation of FFDC on HP Integrity NonStop™ NS-series servers.
This section provides a high-level introduction to TFDS on Integrity NonStop NS-series
servers and addresses these topics:
•
Overview on page 1-1
•
Architecture on page 1-2
•
Functionality on page 1-4
•
Installation on page 1-8
Overview
TFDS is a key component of HP’s problem management system for Integrity NonStop
NS-series servers. TFDS automates many tasks associated with data collection and
resource recovery in the event of software, subsystem, or processor failure.
In response to a failure, TFDS collects certain information about the failure and
compares it to existing information in the TFDS incident database to determine
whether the failure is a first-time occurrence or the result of a recurring defect.
For first-time failure occurrences, TFDS records the new failure in the incident
database and gathers additional information required for problem identification and
resolution during subsequent analysis and repair efforts. (For more information, see
Files Included in Failure Data Collection on page 1-7.) For recurring failures, however,
only the recurrence is recorded in the incident database; redundant efforts to identify
and analyze the failure are avoided.
In the event that a failure results in a processor halt, TFDS also reloads and dumps the
halted processor automatically, eliminating downtime associated with manual data
collection and processor reloading. For recurring failures, TFDS also suppresses
redundant processor dumps, saving potentially large amounts of disk space.
In addition to saving time, money, and resources in identifying and solving problems
associated with failures, TFDS is also flexible, allowing you to configure it to meet your
specific needs. Configuration options include automated dumping and reloading of
processors, processor dump placement, and dump file analysis. (For more information,
see Configuring TFDS on page 2-7.)