<div class="content" name="Overview (synopsis)" uuid="ce2936df-6147-4c12-bda3-72c51d64d593"><p>In complex distributed systems, a situation can arise where
an error encountered on one network node has to be communicated to another
network node. A protocol that is used to transmit data between network nodes
usually has some provisions to transmit errors in its messages, but often the
error that is being communicated is a single unsigned integer or a single
unsigned integer plus a short string. As the complexity of the system and/or
the number of network nodes that are involved grows, a single unsigned integer
and/or a short string might prove insufficient for quick and efficient
troubleshooting of all possible scenarios.</p><p>This specification defines an encoding for a rich,
structured error called an extended error. After the extended error is encoded,
it has to be transmitted between network nodes by a protocol outside this
specification.</p><p>The extended error itself is used for troubleshooting a
malfunctioning system and is intended to be used by a human reader or an
automated failure diagnostic system. This specification does not prescribe what
the extended error should be; it specifies the fields and field values that are
used for encoding the extended error (see section <a href="a15bedb2-563c-4934-b3de-280084ae1e14" data-linktype="relative-path">2</a>). Protocols and systems
are free to create and encode any extended error a support engineer or an
expert user of the system might find useful to troubleshoot a malfunctioning
system.</p></div>