<?xml version="1.0" encoding="utf-8" ?>
<!DOCTYPE chapter SYSTEM "chapter.dtd">
<chapter>
<header>
<copyright>
<year>1997</year><year>2013</year>
<holder>Ericsson AB. All Rights Reserved.</holder>
</copyright>
<legalnotice>
The contents of this file are subject to the Erlang Public License,
Version 1.1, (the "License"); you may not use this file except in
compliance with the License. You should have received a copy of the
Erlang Public License along with this software. If not, it can be
retrieved online at http://www.erlang.org/.
Software distributed under the License is distributed on an "AS IS"
basis, WITHOUT WARRANTY OF ANY KIND, either express or implied. See
the License for the specific language governing rights and limitations
under the License.
</legalnotice>
<title>Agent Functional Description</title>
<prepared></prepared>
<responsible></responsible>
<docno></docno>
<approved></approved>
<checked></checked>
<date></date>
<rev></rev>
<file>snmp_agent_funct_descr.xml</file>
</header>
<p>The SNMP agent system consists of one Master Agent and
optional Sub-agents.
</p>
<p>The tool makes it easy to dynamically extend an SNMP agent in
run-time. MIBs can be loaded and unloaded at any time. It is also
easy to change the implementation of an MIB in run-time, without
having to recompile the MIB. The MIB implementation is clearly
separated from the agent.
</p>
<p>To facilitate incremental MIB implementation, the tool can
generate a prototype implementation for a whole MIB, or parts
thereof. This allows different MIBs and management applications to
be developed at the same time.
</p>
<section>
<title>Features</title>
<marker id="features"></marker>
<p>To implement an agent, the programmer writes instrumentation
functions for the variables and the tables in the MIBs that the
agent is going to support. A running prototype which handles <c>set</c>,
<c>get</c>, and <c>get-next</c> can be created without any programming.
</p>
<p>The toolkit provides the following:
</p>
<list type="bulleted">
<item>multi-lingual multi-threaded extensible SNMP agent</item>
<item>easy writing of instrumentation functions with a
high-level programming language</item>
<item>basic fault handling such as automatic type checking</item>
<item>access control</item>
<item>authentication</item>
<item>privacy through encryption</item>
<item>loading and unloading of MIBs in run-time</item>
<item>the ability to change instrumentation functions without
recompiling the MIB</item>
<item>rapid prototyping environment where the MIB compiler can
use generic instrumentation functions, which later can be
refined by the programmer</item>
<item>a simple and extensible model for transaction handling and
consistency checking of set-requests</item>
<item>support of the sub-agent concept via distributed Erlang</item>
<item>a mechanism for sending notifications (traps and informs)</item>
<item>support for implementing SNMP tables in the Mnesia DBMS.</item>
</list>
</section>
<section>
<title>SNMPv1, SNMPv2 and SNMPv3</title>
<marker id="versions"></marker>
<p>The SNMP development toolkit works with all three versions of
Standard Internet Management Framework; SNMPv1, SNMPv2 and SNMPv3.
They all share the same basic structure and components. And they
follow the same architecture.</p>
<p>The versions are defined in following RFCs</p>
<list type="bulleted">
<item>SNMPv1 RFC 1555, 1157 1212, 1213 and 1215</item>
<item>SNMPv2 RFC 1902 - 1907</item>
<item>SNMPv3 RFC 2570 - 2575</item>
</list>
<p>Over time, as the Framework has evolved from SNMPv1 , through SNMPv2,
to SNMPv3 the definitions of each of these architectural components
have become richer and more clearly defined, but the fundamental
architecture has remained consistent.</p>
<p>The main features of SNMPv2 compared to SNMPv1 are:
</p>
<list type="bulleted">
<item>The <c>get-bulk</c> operation for transferring large
amounts of data.
</item>
<item>Enhanced error codes.
</item>
<item>A more precise language for MIB specification</item>
</list>
<p>The standard documents that define SNMPv2 are incomplete, in
the sense that they do not specify how an SNMPv2 message looks
like. The message format and security issues are left to a
special Administrative Framework. One such framework is the
Community-based SNMPv2 Framework (SNMPv2c), which uses the same
message format and framework as SNMPv1. Other
experimental frameworks as exist, e.g. SNMPv2u and SNMPv2*.
</p>
<p>The SNMPv3 specifications take a modular
approach to SNMP. All modules are
separated from each other, and can be extended or replaced
individually. Examples of modules are Message definition,
Security and Access Control. The main features of SNMPv3 are:
</p>
<list type="bulleted">
<item>Encryption and authentication is added.
</item>
<item>MIBs for agent configuration are defined.</item>
</list>
<p>All these specifications are commonly referred to as "SNMPv3",
but it is actually only the Message module, which defines a new
message format, and Security module, which takes care of
encryption and authentication, that cannot be used with SNMPv1 or
SNMPv2c. In this version of the agent toolkit, all the standard
MIBs for agent configuration are used. This includes MIBs for
definition of management targets for notifications. These MIBs
are used regardless of which SNMP version the agent is configured
to use.
</p>
<p>The extensible agent in this toolkit understands the SNMPv1,
SNMPv2c and SNMPv3. Recall that SNMP consists of two separate
parts, the MIB definition language (SMI), and the protocol. On
the protocol level, the agent can be configured to speak v1, v2c,
v3 or any combination of them at the same time, i.e. a v1 request
gets an v1 reply, a v2c request gets a v2c reply, and a v3 request
gets a v3 reply. On the MIB level, the MIB compiler can compile
both SMIv1 and SMIv2 MIBs. Once compiled, any of the formats can
be loaded into the agent, regardless of which protocol version the
agent is configured to use. This means that the agent translates
from v2 notifications to v1 traps, and vice versa. For example,
v2 MIBs can be loaded into an agent that speaks v1 only. The
procedures for the translation between the two protocols are
described in RFC 1908 and RFC 2089.
</p>
<p>In order for an implementation to make full use of the enhanced
SNMPv2 error codes, it is essential that the instrumentation
functions always return SNMPv2 error codes, in case of error.
These are translated into the corresponding SNMPv1 error codes by
the agent, if necessary.</p>
<note>
<p>The translation from an SMIv1 MIB to an SNMPv2c or SNMPv3 reply
is always very straightforward, but the translation from a v2 MIB
to a v1 reply is somewhat more complicated. There is one data
type in SMIv2, called <c>Counter64</c>, that an SNMPv1 manager cannot
decode correctly. Therefore, an agent may never send a <c>Counter64</c>
object to an SNMPv1 manager. The common practice in these
situations is to simple ignore any <c>Counter64</c> objects, when sending
a reply or a trap to an SNMPv1 manager. For example, if an SNMPv1
manager tries to GET an object of type <c>Counter64</c>, he will get a
<c>noSuchName</c> error, while an SNMPv2 manager would get a
correct value.</p>
</note>
</section>
<section>
<title>Operation</title>
<marker id="operations"></marker>
<p>The following steps are needed to get a running agent:</p>
<list type="ordered">
<item>
<p>Write your MIB in SMI in a text file.</p>
</item>
<item>
<p>Write the instrumentation functions in Erlang and compile them.</p>
</item>
<item>
<p>Put their names in the association file.</p>
</item>
<item>
<p>Run the MIB together with the association file through the
MIB compiler.</p>
</item>
<item>
<p>Configure the application (agent).</p>
</item>
<item>
<p>Start the application (agent).</p>
</item>
<item>
<p>Load the compiled MIB into the agent.</p>
</item>
</list>
<p>The figures in this section illustrate the steps involved in
the development of an SNMP agent.</p>
<marker id="image-1"></marker>
<image file="snmp-um-1-image-1.gif">
<icaption>MIB Compiler Principles</icaption>
</image>
<p>The compiler parses the SMI file and associates each table or
variable with an instrumentation function (see the figure <seealso marker="#image-1">MIB Compiler Principles</seealso>). The actual
instrumentation functions are not needed at MIB compile time, only
their names.
</p>
<p>The binary output file produced by the compiler is read by the
agent at MIB load time (see the figure <seealso marker="#image-2">Starting the Agent</seealso>). The instrumentation is ordinary Erlang code which
is loaded explicitly or automatically the first time it is called.</p>
<marker id="image-2"></marker>
<image file="snmp-um-1-image-2.gif">
<icaption>Starting the Agent</icaption>
</image>
<p>The SNMP agent system consists of one Master Agent and optional
sub-agents. The Master Agent can be seen as a special kind of
sub-agent. It implements the core agent functionality, UDP packet
processing, type checking, access control, trap distribution, and
so on. From a user perspective, it is used as an ordinary
sub-agent.
</p>
<p>Sub-agents are only needed if your application requires special
support for distribution from the SNMP toolkit. A sub-agent can
also be used if the application requires a more complex set
transaction scheme than is found in the master agent.
</p>
<p>The following illustration shows how a system can look in runtime.</p>
<marker id="snmp_ch2_fig3"></marker>
<image file="snmp-um-1-image-3.gif">
<icaption>Architecture</icaption>
</image>
<p>A typical operation could include the following steps:</p>
<list type="ordered">
<item>The Manager sends a request to the Agent.</item>
<item>The Master Agent decodes the incoming UDP packet.</item>
<item>The Master Agent determines which items in the request
that should be processed here and which items should be
forwarded to its subagent.</item>
<item>Step 3 is repeated by all subagents.</item>
<item>Each sub-agent calls the instrumentation for its loaded MIBs.</item>
<item>The results of calling the instrumentation are propagated
back to the Master Agent.</item>
<item>The answer to the request is encoded to a UDP Protocol
Data Unit (PDU).</item>
</list>
<p>The sequence of steps shown is probably more complex than
normal, but it illustrates the amount of functionality which is
available. The following points should be noted:
</p>
<list type="bulleted">
<item>An agent can have many MIBs loaded at the same time.</item>
<item>Sub-agents can also have sub-agents. Each sub-agent can have
an arbitrary number of child sub-agents registered, forming a
hierarchy.</item>
<item>One MIB can communicate with many applications.</item>
<item>Instrumentation can use Distributed Erlang to communicate
with an application.</item>
</list>
<p>Most applications only need the Master Agent because an agent
can have multiple MIBs loaded at the same time.</p>
</section>
<section>
<title>Sub-agents and MIB Loading</title>
<marker id="sub_agent_mib_loading"></marker>
<p>Since applications tend to be transient (they are dynamically
loaded and unloaded), the management of these applications must be
dynamic as well. For example, if we have an equipment MIB for a
rack and different MIBs for boards, which can be installed in the
rack, the MIB for a card should be loaded when the card is
inserted, and unloaded when the card is removed.
</p>
<p>In this agent system, there are two ways to dynamically install
management information. The most common way is to load an MIB into
an agent. The other way is to use a sub-agent, which is controlled
by the application and is able to register and unregister itself. A
sub-agent can register itself for managing a sub-tree (not to be mixed up
with <c>erlang:register</c>). The sub-tree is identified by an
Object Identifier. When a sub-agent is registered, it receives all
requests for this particular sub-tree and it is responsible for
answering them. It should also be noted that a sub-agent can be
started and stopped at any time.
</p>
<p>Compared to other SNMP agent packages, there is a significant
difference in this way of using sub-agents. Other packages normally
use sub-agents to load and unload MIBs in run-time. In Erlang, it is
easy to load code in run-time and it is possible to load an MIB
into an existing sub-agent. It is not necessary to create a new process
for handling a new MIB.
</p>
<p>Sub-agents are used for the following reasons:
</p>
<list type="bulleted">
<item>to provide a more complex set-transaction scheme than
master agent</item>
<item>to avoid unnecessary process communication</item>
<item>to provide a more lightweight mechanism for loading and
unloading MIBs in run-time</item>
<item>to provide interaction with other SNMP agent toolkits.</item>
</list>
<p>Refer to the chapter
<seealso marker="snmp_advanced_agent">Advanced Agent Topics</seealso>
in this User's Guide for more information about these topics.
</p>
<p>The communication protocol between sub-agents is the normal
message passing which is used in distributed Erlang systems. This
implies that sub-agent communication is very efficient compared to
SMUX, DPI, AgentX, and similar protocols.</p>
</section>
<section>
<title>Contexts and Communities</title>
<marker id="context_and_communities"></marker>
<p>A context is a collection of management information accessible
by an SNMP entity. An instance of a management object may exist in
more than one context. An SNMP entity potentially has access to
many contexts.</p>
<p>Each managed object can exist in many instances within a
SNMP entity. To identify the instances, specified by an MIB module,
a method to distinguish the actual instance by its 'scope' or
context is used. Often the context is a physical or a logical device.
It can include multiple devices, a subset of a single device or a
subset of multiple devices, but the context is always
defined as a subset of a single SNMP entity. To be able to
identify a specific
item of management information within an SNMP entity, the context,
the object type and its instance must be used.</p>
<p>For example, the managed object type <c>ifDescr</c> from RFC1573, is
defined as the description of a network interface. To identify
the description of device-X's first network interface, four pieces
of information are needed: the snmpEngineID of the SNMP entity
which provides access to the management information at device-X,
the <c>contextName</c> (device-X), the managed object type
(<c>ifDescr</c>), and the instance ("1").
</p>
<p>In SNMPv1 and SNMPv2c, the community string in the message was
used for (at least) three different purposes:
</p>
<list type="bulleted">
<item>
<p>to identify the context</p>
</item>
<item>
<p>to provide authentication</p>
</item>
<item>
<p>to identify a set of trap targets</p>
</item>
</list>
<p>In SNMPv3, each of these usage areas has its own unique
mechanism. A context is identified by the name of the SNMP
entity, <c>contextEngineID</c>, and the name of the context,
<c>contextName</c>. Each SNMPv3 message contains values for these
two parameters.
</p>
<p>There is a MIB, SNMP-COMMUNITY-MIB, which maps a community
string to a <c>contextEngineID</c> and <c>contextName</c>. Thus,
each message, an SNMPv1, SNMPv2c or an SNMPv3 message, always
uniquely identifies a context.
</p>
<p>For an agent, the <c>contextEngineID</c> identified by a received
message, is always equal to the <c>snmpEngineID</c> of the agent.
Otherwise, the message was not intended for the agent. If the
agent is configured with more than one context, the
instrumentation code must be able to figure out for which context
the request was intended. There is a function
<c>snmpa:current_context/0</c> provided for this purpose.
</p>
<p>By default, the agent has no knowledge of any other contexts
than the default context, <c>""</c>. If it is to support more
contexts, these must be explicitly added, by using an appropriate
configuration file
<seealso marker="snmp_agent_config_files">Agent Configuration Files</seealso>.
</p>
</section>
<section>
<title>Management of the Agent</title>
<marker id="management"></marker>
<p>There is a set of standard MIBs, which are used to control and
configure an SNMP agent. All of these MIBs, with the exception of
the optional SNMP-PROXY-MIB (which is only used for proxy agents),
are implemented in this agent. Further, it is configurable which
of these MIBs are actually loaded, and thus made visible to SNMP
managers. For example, in a non-secure environment, it might be a
good idea to not make MIBs that define access control visible.
Note, the data the MIBs define is used internally in the
agent, even if the MIBs not are loaded. This chapter describes
these standard MIBs, and some aspects of their implementation.
</p>
<p>Any SNMP agent must implement the <c>system</c> group and the
<c>snmp</c> group, defined in MIB-II. The definitions of these
groups have changed from SNMPv1 to SNMPv2. MIBs and implementations
for both of these versions are Provided in the
distribution. The MIB file for SNMPv1 is called STANDARD-MIB, and the
corresponding for SNMPv2 is called SNMPv2-MIB. If the agent is
configured for SNMPv1 only, the STANDARD-MIB is loaded by default;
otherwise, the SNMPv2-MIB is loaded by default. It is possible to
override this default behavior, by explicitly loading another
version of this MIB, for example, you could choose to implement
the union of all objects in these two MIBs.
</p>
<p>An SNMPv3 agent must implement the SNMP-FRAMEWORK-MIB and
SNMP-MPD-MIB. These MIBs are loaded by default, if the agent is
configured for SNMPv3. These MIBs can be loaded for other
versions as well.
</p>
<p>There are five other standard MIBs, which also may be loaded
into the agent. These MIBs are:
</p>
<list type="bulleted">
<item>
<p>SNMP-TARGET-MIB and SNMP-NOTIFICATION-MIB, which defines
managed objects for configuration of management targets,
i.e. receivers of notifications (traps and informs). These
MIBs can be used with any SNMP version.
</p>
</item>
<item>
<p>SNMP-VIEW-BASED-ACM-MIB, which defined managed objects
for access control. This MIB can be used with any SNMP
version.
</p>
</item>
<item>
<p>SNMP-COMMUNITY-MIB, which defines managed objects for
coexistence of SNMPv1 and SNMPv2c with SNMPv3. This MIB is
only useful if SNMPv1 or SNMPv2c is used, possibly in
combination with SNMPv3.
</p>
</item>
<item>
<p>SNMP-USER-BASED-SM-MIB, which defines managed objects
for authentication and privacy. This MIB is only useful
with SNMPv3.
</p>
</item>
</list>
<p>All of these MIBs should be loaded into the Master Agent. Once
loaded, these MIBs are always available in all contexts.
</p>
<p>The ASN.1 code, the Erlang source code, and the generated
<c>.hrl</c> files for them are provided in the distribution and are
placed in the directories <c>mibs</c>, <c>src</c>, and <c>include</c>,
respectively, in the <c>snmp</c> application.
</p>
<p>The <c>.hrl</c> files are generated with
<c>snmpc:mib_to_hrl/1</c>. Include these files in your code as in
the following example:
</p>
<code type="none">
-include_lib("snmp/include/SNMPv2-MIB.hrl").
</code>
<p>The initial values for the managed objects defined in these
tables, are read at start-up from a set of configuration files.
These are described in <seealso marker="snmp_config">Configuration Files</seealso>.
</p>
<section>
<title>STANDARD-MIB and SNMPv2-MIB</title>
<p>These MIBs contain the <c>snmp-</c> and <c>system</c> groups
from MIB-II which is defined in RFC1213 (STANDARD-MIB) or
RFC1907 (SNMPv2-MIB). They are implemented in the
<c>snmp_standard_mib</c> module. The <c>snmp</c> counters all
reside in volatile memory and the <c>system</c> and
<c>snmpEnableAuthenTraps</c> variables in persistent memory,
using the SNMP built-in database (refer to the Reference Manual,
section <c>snmp</c>, module <c>snmpa_local_db</c> for more
details).</p>
<p>If another implementation of any of these variables is needed,
e.g. to store the persistent variables in a Mnesia database,
an own implementation of the variables must be made. That MIB
will be compiled and loaded instead of the default MIB.
The new compiled MIB
must have the same name as the original MIB (i.e. STANDARD-MIB
or SNMPv2-MIB), and be located in the SNMP configuration
directory (see <seealso marker="snmp_config">Configuration Files</seealso>.)
</p>
<p>One of these MIBs is always loaded. If only SNMPv1 is used,
STANDARD-MIB is loaded, otherwise SNMPv2-MIB is loaded.
</p>
<section>
<title>Data Types</title>
<p>There are some new data types in SNMPv2 that are useful in
SNMPv1 as well. In the STANDARD-MIB, three data types are
defined, <c>RowStatus</c>, <c>TruthValue</c> and
<c>DateAndTime</c>. These data types are originally defined
as textual conventions in SNMPv2-TC (RFC1903).
</p>
</section>
</section>
<section>
<title>SNMP-FRAMEWORK-MIB and SNMP-MPD-MIB</title>
<p>The SNMP-FRAMEWORK-MIB and SNMP-MPD-MIB define additional
read-only managed objects, which
is used in the generic SNMP framework defined in RFC2271 and the
generic message processing and dispatching module defined in
RFC2272. They are generic in the sense that they are not tied
to any specific SNMP version.
</p>
<p>The objects in these MIBs are implemented in the modules
<c>snmp_framework_mib</c> and <c>snmp_standard_mib</c>,
respectively. All objects reside in volatile memory, and the
configuration files are always reread at start-up.
</p>
<p>If SNMPv3 is used, these MIBs are loaded by default.
</p>
</section>
<section>
<title>SNMP-TARGET-MIB and SNMP-NOTIFICATION-MIB</title>
<p>The SNMP-TARGET-MIB and SNMP-NOTIFICATION-MIB define managed
objects for configuration of notification receivers. They
are described in detail in RFC2273. Only a brief description
is given here.
</p>
<p>All tables in these MIBs have a column of type
<c>StorageType</c>. The value of this column specifies how each
row is stored, and what happens in case of a restart of the
agent. The implementation supports the values <c>volatile</c>
and <c>nonVolatile</c>. When the tables are initially filled
with data from the configuration files, these rows will
automatically have storage type <c>nonVolatile</c>. Should the
agent restart, all <c>nonVolatile</c> rows survive the restart,
while the <c>volatile</c> rows are lost.
The configuration files are not read at restart, by default.
</p>
<p>These MIBs are not loaded by default.
</p>
<section>
<title>snmpNotifyTable</title>
<p>An entry in the <c>snmpNotifyTable</c> selects a set
of management targets, which should receive notifications,
as well as the type (trap or inform) of notification that
should be sent to each selected management target.
When an application sends a notification using
the function <c>send_notification/5</c> or the function
<c>send_trap</c> the parameter <c>NotifyName</c>, specified in
the call, is used as an index in the table. The notification
is sent to the management targets selected by that entry.
</p>
</section>
<section>
<title>snmpTargetAddrTable</title>
<p>An entry in the <c>snmpTargetAddrTable</c> defines
transport parameters (such as
IP address and UDP port) for each management target. Each row
in the <c>snmpNotifyTable</c> refers to potentially many rows
in the <c>snmpTargetAddrTable</c>. Each row in the
<c>snmpTargetAddrTable</c>
refers to an entry in the <c>snmpTargetParamsTable</c>.
</p>
</section>
<section>
<title>snmpTargetParamsTable</title>
<p>An entry in the <c>snmpTargetParamsTable</c> defines
which SNMP version to use, and which security parameters to use.
</p>
<p>Which SNMP version to use is implicitly defined by
specifying the Message Processing Model. This version of the
agent handles the models <c>v1</c>, <c>v2c</c> and <c>v3</c>.
</p>
<p>Each row specifies which security model to use, along with
security level and security parameters.
</p>
</section>
</section>
<section>
<title>SNMP-VIEW-BASED-ACM-MIB</title>
<p>The SNMP-VIEW-BASED-ACM-MIB defines managed objects to
control access to the the managed objects for the managers.
The View Based Access Control Module (VACM) can be used with
any SNMP version. However, if it is used with SNMPv1 or SNMPv2c,
the SNMP-COMMUNITY-MIB defines additional objects to map
community strings to VACM parameters.
</p>
<p>All tables in this MIB have a column of type <c>StorageType</c>.
The value of this column specifies how each
row is stored, and what happens in case of a restart of the
agent. The implementation supports the values <c>volatile</c>
and <c>nonVolatile</c>. When the tables are initially filled
with data from the configuration files, these rows will
automatically have storage type <c>nonVolatile</c>. Should the
agent restart, all <c>nonVolatile</c> rows survive the restart,
while the <c>volatile</c> rows are lost.
The configuration files are not read at restart by default.
</p>
<p>This MIB is not loaded by default.
</p>
<p>VACM is described in detail in RFC2275. Here is only a brief
description given.
</p>
<p>The basic concept is that of a <em>MIB view</em>. An MIB view
is a subset of all the objects implemented by an agent. A
manager has access to a certain MIB view, depending on which
security parameters are used, in which context the request is
made, and which type of request is made.
</p>
<p>The following picture gives an overview of the mechanism to
select an MIB view:</p>
<image file="MIB_mechanism.gif">
<icaption>Overview of the mechanism of MIB selection</icaption>
</image>
<section>
<title>vacmContextTable</title>
<p>The <c>vacmContextTable</c> is a read-only table that lists all
available contexts.
</p>
</section>
<section>
<title>vacmSecurityToGroupTable</title>
<p>The <c>vacmSecurityToGroupTable</c> maps a <c>securityModel</c>
and a
<c>securityName</c> to a <c>groupName</c>.
</p>
</section>
<section>
<title>vacmAccessTable</title>
<p>The <c>vacmAccessTable</c> maps the <c>groupName</c> (found in
<c>vacmSecurityToGroupTable</c>), <c>contextName</c>,
<c>securityModel</c>, and <c>securityLevel</c> to an MIB view
for each type of operation (read, write, or notify). The MIB
view is represented as a <c>viewName</c>. The definition of
the MIB view represented by the <c>viewName</c> is found in
the <c>vacmViewTreeFamilyTable</c></p>
</section>
<section>
<title>vacmViewTreeFamilyTable</title>
<p>The <c>vacmViewTreeFamilyTable</c> is indexed by the
<c>viewName</c>, and defines
which objects are included in the MIB view.
</p>
<p>The MIB definition for the table looks as follows:</p>
<pre>
VacmViewTreeFamilyEntry ::= SEQUENCE
{
vacmViewTreeFamilyViewName SnmpAdminString,
vacmViewTreeFamilySubtree OBJECT IDENTIFIER,
vacmViewTreeFamilyMask OCTET STRING,
vacmViewTreeFamilyType INTEGER,
vacmViewTreeFamilyStorageType StorageType,
vacmViewTreeFamilyStatus RowStatus
}
INDEX { vacmViewTreeFamilyViewName,
vacmViewTreeFamilySubtree
}
</pre>
<p>Each <c>vacmViewTreeFamilyViewName</c> refers to a
collection of sub-trees.
</p>
<section>
<title>MIB View Semantics</title>
<p>An MIB view is a collection of included and excluded
sub-trees. A sub-tree is identified by an OBJECT IDENTIFIER. A
mask is associated with each sub-tree.
</p>
<p>For each possible MIB object instance, the instance
belongs to a sub-tree if:
</p>
<list type="bulleted">
<item>the OBJECT IDENTIFIER name of that MIB object
instance comprises at least as many sub-identifiers as
does the sub-tree, and
</item>
<item>each sub-identifier in the name of that MIB object
instance matches the corresponding sub-identifier of the
sub-tree whenever the corresponding bit of the associated
mask is 1 (0 is a wild card that matches anything).</item>
</list>
<p>Membership of an object instance in an MIB view is
determined by the following algorithm:
</p>
<list type="bulleted">
<item>If an MIB object instance does not belong to any of
the relevant sub-trees, then the instance is not in the
MIB view.
</item>
<item>If an MIB object instance belongs to exactly one
sub-tree, then the instance is included in, or excluded
from, the relevant MIB view according to the type of
that entry.
</item>
<item>If an MIB object instance belongs to more than one
sub-tree, then the sub-tree which comprises the greatest
number of sub-identifiers, and is the lexicographically
greatest, is used.
</item>
</list>
<note>
<p>If the OBJECT IDENTIFIER is longer than an OBJECT
IDENTIFIER of an object type in the MIB, it refers to
object instances. Because of this, it is possible to
control whether or not particular rows in a table shall be
visible.</p>
</note>
</section>
</section>
</section>
<section>
<title>SNMP-COMMUNITY-MIB</title>
<p>The SNMP-COMMUNITY-MIB defines managed objects that is used
for coexistence between SNMPv1 and SNMPv2c with SNMPv3.
Specifically, it contains objects for mapping between community
strings and version-independent SNMP message parameters. In
addition, this MIB provides a mechanism for performing source address
validation on incoming requests, and for selecting community
strings based on target addresses for outgoing notifications.
</p>
<p>All tables in this MIB have a column of type
<c>StorageType</c>. The value of this column specifies how each
row is stored, and what happens in case of a restart of the
agent. The implementation supports the values <c>volatile</c>
and <c>nonVolatile</c>. When the tables are initially filled
with data from the configuration files, these rows will
automatically have storage type <c>nonVolatile</c>. Should the
agent restart, all <c>nonVolatile</c> rows survive the restart,
while the <c>volatile</c> rows are lost.
The configuration files are not read at restart, by default.
</p>
<p>This MIB is not loaded by default.
</p>
</section>
<section>
<title>SNMP-USER-BASED-SM-MIB</title>
<p>The SNMP-USER-BASED-SM-MIB defines managed objects that is
used for the User-Based Security Model.
</p>
<p>All tables in this MIB have a column of type
<c>StorageType</c>. The value of the column specifies how each
row is stored, and what happens in case of a restart of the
agent. The implementation supports the values <c>volatile</c>
and <c>nonVolatile</c>. When the tables are initially filled
with data from the configuration files, these rows will
automatically have storage type <c>nonVolatile</c>. Should the
agent restart, all <c>nonVolatile</c> rows survive the restart,
while the <c>volatile</c> rows are lost.
The configuration files are not read at restart, by default.
</p>
<p>This MIB is not loaded by default.
</p>
</section>
<section>
<title>OTP-SNMPEA-MIB</title>
<p>The OTP-SNMPEA-MIB was used in earlier versions of the agent, before
standard MIBs existed for access control, MIB views, and trap
target specification. All objects in this MIB are now obsolete.
</p>
</section>
</section>
<section>
<title>Notifications</title>
<marker id="notifications"></marker>
<p>Notifications are defined in SMIv1 with the TRAP-TYPE macro in
the definition of an MIB (see RFC1215). The corresponding
macro in SMIv2 is NOTIFICATION-TYPE. When an application
decides to send a notification, it calls one of the following
functions:
</p>
<code type="none">
snmpa:send_notification(Agent, Notification, Receiver
[, NotifyName, ContextName, Varbinds])
snmpa:send_trap(Agent, Notification, Community [, Receiver, Varbinds])
</code>
<p>providing the registered name or process identifier of the
agent where the MIB, which defines the notification is loaded and
the symbolic name of the notification.
</p>
<p>If the <c>send_notification/3,4</c> function is used, all
management targets are selected, as defined in RFC2273. The
<c>Receiver</c> parameter defines where the agent should send
information about the delivery of inform requests.
</p>
<p>If the <c>send_notification/5</c> function is used, an
<c>NotifyName</c> must be provided. This parameter is used as an
index in the <c>snmpNotifyTable</c>, and the management targets
defined by that single entry is used.
</p>
<p>The <c>send_notification/6</c> function is the most general
version of the function. A <c>ContextName</c> must be specified,
from which the notification will be sent. If this parameter is
not specified, the default context (<c>""</c>) is used.
</p>
<p>The function <c>send_trap</c> is kept for backwards
compatibility and should not be used in new code. Applications
that use this function will
continue to work. The <c>snmpNotifyName</c> is used as the
community string by the agent when a notification is sent.
</p>
<section>
<title>Notification Sending</title>
<p>The simplest way to send a notification is to call the function
<c>snmpa:send_notification(Agent, Notification, no_receiver)</c>.
In this case, the agent performs a get-operation to retrieve the
object values that are defined in the notification
specification (with the TRAP-TYPE or NOTIFICATION-TYPE macros).
The notification is sent to all managers defined in the target
and notify tables, either unacknowledged as traps, or
acknowledged as inform requests.
</p>
<p>If the caller of the function wants to know whether or not
acknowledgments are received for a certain notification
(provided it is sent as an inform), the <c>Receiver</c>
parameter can be specified as <c>{Tag, ProcessName}</c> (refer
to the Reference Manual, section snmp, module <c>snmp</c> for
more details). In this case, the agent send a message
<c>{snmp_notification, Tag, {got_response, ManagerAddr}}</c> or
<c>{snmp_notification, Tag, {no_response, ManagerAddr}}</c> for
each management target.
</p>
<p>Sometimes it is not possible to retrieve the values for some
of the objects in the notification specification with a
get-operation. However, they are known when the
<c>send_notification</c> function is called. This is the case if
an object is an element in a table. It is possible to give the
values of some objects to the <c>send_notification</c> function
<c>snmpa:send_notification(Agent, Notification, Receiver, Varbinds)</c>. In this function, <c>Varbinds</c> is a list of
<c>Varbind</c>, where each <c>Varbind</c> is one of:
</p>
<list type="bulleted">
<item><c>{Variable, Value}</c>, where <c>Variable</c> is the
symbolic name of a scalar variable referred to in the notification
specification.
</item>
<item><c>{Column, RowIndex, Value}</c>, where <c>Column</c> is
the symbolic name of a column variable. <c>RowIndex</c> is a
list of indices for the specified element. If this is the
case, the OBJECT IDENTIFIER sent in the trap is the
<c>RowIndex</c> appended to the OBJECT IDENTIFIER for the
table column. This is the OBJECT IDENTIFIER which specifies
the element.
</item>
<item><c>{OID, Value}</c>, where <c>OID</c> is the OBJECT
IDENTIFIER for an instance of an object, scalar variable or
column variable.
</item>
</list>
<p>For example, to specify that <c>sysLocation</c> should have the
value <c>"upstairs"</c> in the notification, we could use one of:
</p>
<list type="bulleted">
<item><c>{sysLocation, "upstairs"}</c> or</item>
<item><c>{[1,3,6,1,2,1,1,6,0], "upstairs"}</c></item>
</list>
<p>It is also possible to specify names and values for extra
variables that should be sent in the notification, but were not
defined in the notification specification.
</p>
<p>The notification is sent to all management targets found in
the tables. However, make sure that each manager has access to
the variables in the notification. If a variable is outside a
manager's MIB view, this manager will not receive the
notification.
</p>
<note>
<p>By definition, it is not possible to send objects with
ACCESS <c>not-accessible</c> in notifications. However,
historically this is often done and for this reason we allow
it in notification sending. If a variable has ACCESS
<c>not-accessible</c>, the user must provide a value for the
variable in the <c>Varbinds</c> list. It is not possible for
the agent to perform a get-operation to retrieve this value.
</p>
</note>
</section>
<section>
<title>Notification Filters</title>
<p>It is possible to add <em>notification filters</em> to an agent.
These filters will be called when a notification is to be
sent. Their purpose is to allow modification, suppression or
other type of actions.</p>
<p>A notification filter is a module implementing the
<seealso marker="snmpa_notification_filter">snmpa_notification_filter</seealso> behaviour. A filter is added/deleted using the functions:
<seealso marker="snmpa#register_notification_filter">snmpa:register_notification_filter</seealso> and
<seealso marker="snmpa#unregister_notification_filter">snmpa:unregister_notification_filter</seealso>.</p>
<p>Unless otherwise specified, the order of the registered filters
will be the order in which they are registered.</p>
</section>
<section>
<title>Sub-agent Path</title>
<p>If a value for an object is not given to the
<c>send_notification</c> function, the sub-agent will perform a
get-operation to retrieve it. If the object is not implemented
in this sub-agent, its parent agent tries to perform a
get-operation to retrieve it. If the object is not implemented
in this agent either, it forwards the object to its parent, and
so on. Eventually the Master Agent is reached and at this point
all unknown object values must be resolved. If some object is
unknown even to the Master Agent, this is regarded as an error
and is reported with a call to <c>user_err/2</c> of the
error report module. No notifications are sent in this case.
</p>
<p>For a given notification, the variables, which are referred to
in the notification specification, must be implemented by the
agent that has the MIB loaded, or by some parent to this
agent. If not, the application must provide values for the
unknown variables. The application must also provide values for
all elements in tables.
</p>
</section>
</section>
<section>
<title>Discovery</title>
<marker id="discovery"></marker>
<p>The <em>sender</em> is <em>authoritative</em> for messages containing
payload which does <em>not</em> expect a response (for example
SNMPv2-Trap, Response or Report PDU). </p>
<p>The <em>receiver</em> is <em>authoritative</em> for messages containing
payload which expects a response (for example
Get, GetNext, Get-Bulk, Set or Inform PDU). </p>
<p>The agent can both perform and respond to discovery.</p>
<p>The agent responds to discovery autonomously, without interaction
by the user. </p>
<p>Initiating discovery towards a manager is done by calling the
<seealso marker="snmpa#discovery">discovery</seealso> function.
The <c>EngineId</c> field of the target (manager) entry in the
<seealso marker="snmp_agent_config_files#target_addr">target_addr.conf</seealso> file has to have the value <c>discovery</c>.
Note that if the manager does not respond, the <c>Timeout</c> and
<c>RetryCount</c>
fields decide how long the function will hang before it returns. </p>
<p>Discovery can only be performed towards one manager at a time.</p>
</section>
</chapter>
