Network Configuration Protocol (NETCONF) is an XML-based network management protocol with filtering capabilities. It provides programmable mechanisms to manage and configure network devices. Through NETCONF, you can configure device parameters, retrieve parameter values, and get statistics information.

In NETCONF messages, each data item is contained in a fixed element. This enables different devices of the same vendor to provide the same access method and the same result presentation method. For the devices of different vendors, XML mapping in NETCONF messages can achieve the same effect. For a network environment containing different devices regardless of vendors, you can develop a NETCONF-based NMS system to configure and manage devices in a simple and effective way.

NETCONF structure

NETCONF has four layers: content layer, operations layer, RPC layer, and transport protocol layer.

Table 1 NETCONF layers and XML layers

NETCONF layer	XML layer	Description
Content	Configuration data, status data, and statistics information	The content layer contains a set of managed objects, which can be configuration data, status data, and statistics information. For information about the operable data, see the NETCONF XML API reference for the device.
Operations	<get>,<get-config>,<edit-config>…	The operations layer defines a set of base operations invoked as RPC methods with XML-encoded parameters. NETCONF base operations include data retrieval operations, configuration operations, lock operations, and session operations. For the device supported operations, see "Supported NETCONF operations."
RPC	<rpc>,<rpc-reply>	The RPC layer provides a simple, transport-independent framing mechanism for encoding RPCs. The <rpc> and <rpc-reply> elements are used to enclose NETCONF requests and responses (data at the operations layer and the content layer).
Transport Protocol	· In non-FIPS mode: Console/Telnet/SSH/HTTP/HTTPS/TLS · In FIPS mode: Console/SSH/HTTPS/TLS	The transport protocol layer provides reliable, connection-oriented, serial data links. In non-FIPS mode, the following login methods are available: · You can log in through Telnet, SSH, or the console port to perform NETCONF operations at the CLI. · You can log in through HTTP or HTTPS to perform NETCONF operations in the Web interface or perform NETCONF-over-SOAP operations. In FIPS mode, all login methods are the same as in non-FIPS mode except that you cannot use HTTP or Telnet.

NETCONF message format

NETCONF

All NETCONF messages are XML-based and comply with RFC 4741. Any incoming NETCONF messages must pass XML Schema check before it can be processed. If a NETCONF message fails XML Schema check, the device sends an error message to the client.

For information about the NETCONF operations supported by the device and the operable data, see the NETCONF XML API reference for the device.

The following example shows a NETCONF message for getting all parameters of all interfaces on the device:

<?xml version="1.0" encoding="utf-8"?>

<get-bulk>

<Ifmgr>

</Interfaces>

</Ifmgr>

</top>

</filter>

</get-bulk>

</rpc>

NETCONF over SOAP

All NETCONF over SOAP messages are XML-based and comply with RFC 4741. NETCONF messages are contained in the <Body> element of SOAP messages. NETCONF over SOAP messages also comply with the following rules:

· SOAP messages must use the SOAP Envelope namespaces.

· SOAP messages must use the SOAP Encoding namespaces.

· SOAP messages cannot contain the following information:

¡ DTD reference.

¡ XML processing instructions.

The following example shows a NETCONF over SOAP message for getting all parameters of all interfaces on the device:

<env:Envelope xmlns:env="http://www.w3.org/2003/05/soap-envelope">

<env:Header>

<auth:Authentication env:mustUnderstand="1" xmlns:auth="http://www.h3c.com/netconf/base:1.0">

<auth:AuthInfo>800207F0120020C</auth:AuthInfo>

</auth:Authentication>

</env:Header>

<env:Body>

<get-bulk>

<Ifmgr>

</Interfaces>

</Ifmgr>

</top>

</filter>

</get-bulk>

</rpc>

</env:Body>

</env:Envelope>

How to use NETCONF

You can use NETCONF to manage and configure the device by using the methods in Table 2.

Table 2 NETCONF methods for configuring the device

Configuration tool	Login method	Remarks
CLI	· Console port · SSH · Telnet	To implement NETCONF operations, copy valid NETCONF messages to the CLI in XML view. This method is suitable for R&D and test purposes.
Standard Web interface for the device	· In FIPS mode: HTTPS · In non-FIPS mode: ¡ HTTP ¡ HTTPS	The system automatically converts the configuration operations on the Web interface to NETCONF messages and sends them to the device to implement NETCONF operations. This method is the most commonly used method.
Custom Web interface	N/A	To use this method, you must enable NETCONF over SOAP. By default, the device cannot interpret Custom Web interfaces' URLs. For the device to interpret these URLs, you must encode the NETCONF messages sent from a custom Web interface in SOAP.

Protocols and standards

· RFC 3339, Date and Time on the Internet: Timestamps

· RFC 4741, NETCONF Configuration Protocol

· RFC 4742, Using the NETCONF Configuration Protocol over Secure SHell (SSH)

· RFC 4743, Using NETCONF over the Simple Object Access Protocol (SOAP)

· RFC 5277, NETCONF Event Notifications

· RFC 5381, Experience of Implementing NETCONF over SOAP

· RFC 5539, NETCONF over Transport Layer Security (TLS)

· RFC 6241, Network Configuration Protocol

FIPS compliance

The device supports the FIPS mode that complies with NIST FIPS 140-2 requirements. Support for features, commands, and parameters might differ in FIPS mode (see Security Configuration Guide) and non-FIPS mode.

NETCONF configuration task list

Tasks at a glance
(Optional.) Configuring NETCONF over SOAP
(Optional.) Enabling NETCONF over SSH
(Optional.) Enabling NETCONF logging
(Optional.) Configuring NETCONF to use module-specific namespaces
(Required.) Establishing a NETCONF session
(Optional.) Subscribing to event notifications
(Optional.) Locking/unlocking the configuration
(Optional.) Performing the <get>/<get-bulk> operation
(Optional.) Performing the <get-config>/<get-bulk-config> operation
(Optional.) Performing the <edit-config> operation
(Optional.) Saving, rolling back, and loading the configuration
(Optional.) Filtering data
(Optional.) Performing CLI operations through NETCONF
(Optional.) Retrieving NETCONF information
(Optional.) Retrieving YANG file content Retrieving YANG file content
(Optional.) Retrieving NETCONF session information
(Optional.) Terminating another NETCONF session
(Optional.) Returning to the CLI

Configuring NETCONF over SOAP

NETCONF over SOAP encapsulates NETCONF messages into SOAP messages and transmits the SOAP messages over HTTP or HTTPS. You can use a custom user interface to establish a NETCONF over SOAP session to the device and perform NETCONF operations.

To configure NETCONF over SOAP:

Step	Command	Remark
1. Enter system view.	system-view	N/A
2. Enable NETCONF over SOAP.	· Enable NETCONF over SOAP over HTTP (not available in FIPS mode): netconf soap http enable · Enable NETCONF over SOAP over HTTPS: netconf soap https enable	By default, the NETCONF over SOAP feature is disabled.
3. Set the DSCP value for NETCONF over SOAP packets.	· Set the DSCP value for NETCONF over SOAP over HTTP packets: netconf soap http dscp dscp-value · Set the DSCP value for NETCONF over SOAP over HTTPs packets: netconf soap https dscp dscp-value	By default, the DSCP value is 0 for NETCONF over SOAP packets.
4. Apply an ACL to control NETCONF over SOAP access.	· Apply an ACL to control NETCONF over SOAP over HTTP access (not available in FIPS mode): netconf soap http acl { acl-number \| name acl-name } · Apply an ACL to control NETCONF over SOAP over HTTPS access: netconf soap https acl { acl-number \| name acl-name }	By default, no ACL is applied to control NETCONF over SOAP access.
5. Specify a mandatory authentication domain for NETCONF users.	netconf soap domain domain-name	By default, no mandatory authentication domain is specified for NETCONF users. For information about authentication domains, see Security Configuration Guide.

Enabling NETCONF over SSH

This feature allows users to use a client to perform NETCONF operations on the device through a NETCONF over SSH connection.

To enable NETCONF over SSH:

Step	Command	Remark
1. Enter system view.	system-view	N/A
2. Enable NETCONF over SSH.	netconf ssh server enable	By default, NETCONF over SSH is disabled.
3. Specify a port to listen for NETCONF over SSH connections.	netconf ssh server port port-number	By default, port 830 listens for NETCONF over SSH connections.

Enabling NETCONF logging

NETCONF logging generates logs for different NETCONF operation sources and NETCONF operations.

To enable NETCONF logging:

Step	Command	Remarks
1. Enter system view.	system-view	N/A
2. Enable NETCONF logging.	netconf log source { all \| { agent \| soap \| web } * } { protocol-operation { all \| { action \| config \| get \| set \| session \| syntax \| others } * } \| row-operation \| verbose }	By default, NETCONF logging is disabled.

Configuring NETCONF to use module-specific namespaces

About module-specific namespaces for NETCONF

NETCONF supports the following types of namespaces:

· Common namespace—The common namespace is shared by all modules. In a packet that uses the common namespace, the namespace is indicated in the <top> element, and the modules are listed under the <top> element.

Example:

<get-bulk>

<Ifmgr>

</Interfaces>

</Ifmgr>

</top>

</filter>

</get-bulk>

</rpc>

· Module-specific namespace—Each module has its own namespace. A packet that uses a module-specific namespace does not have the <top> element. The namespace follows the module name.

Example:

<get-bulk>

</Interfaces>

</Ifmgr>

</filter>

</get-bulk>

</rpc>

The common namespace is incompatible with module-specific namespaces. To set up a NETCONF session, the device and the client must use the same type of namespaces. By default, the common namespace is used. If the client does not support the common namespace, use this feature to configure the device to use module-specific namespaces.

Configuration restrictions and guidelines

For this feature to take effect, you must reestablish the NETCONF session.

Configuration procedure

To configure NETCONF to use module-specific namespaces:

Step	Command	Remark
1. Enter system view.	system-view	N/A
2. Configure NETCONF to use module-specific namespaces.	netconf capability specific-namespace	By default, the common namespace is used.

Establishing a NETCONF session

After a NETCONF session is established, the device automatically sends its capabilities to the client. You must send the capabilities of the client to the device before you can perform any other NETCONF operations.

You can use the aaa session-limit command to set the maximum number of NETCONF sessions that the device can support. If the upper limit is reached, new NETCONF users cannot access the device. For information about this command, see Security Configuration Guide.

Before performing a NETCONF operation, make sure no other users are configuring or managing the device. If multiple users simultaneously configure or manage the device, the result might be different from what you expect.

Setting the NETCONF session idle timeout time

If no packets are exchanged between the device and a user within the NETCONF session idle timeout time, the device tears down the session.

To set the NETCONF session idle timeout time:

Task	Command	Remarks
1. Enter system view.	system-view	N/A
2. Set the NETCONF session idle timeout time.	netconf { soap \| agent } idle-timeout minute	By default, the NETCONF session idle timeout time is as follows: · 10 minutes for NETCONF over SOAP over HTTP sessions and NETCONF over SOAP over HTTPS sessions. · 0 minutes for SSH sessions, Telnet sessions, and NETCONF over SSH sessions.

Task

Command

Remarks

1. Enter system view.

system-view

N/A

2. Set the NETCONF session idle timeout time.

netconf { soap | agent } idle-timeout minute

By default, the NETCONF session idle timeout time is as follows:

· 10 minutes for NETCONF over SOAP over HTTP sessions and NETCONF over SOAP over HTTPS sessions.

· 0 minutes for SSH sessions, Telnet sessions, and NETCONF over SSH sessions.

Entering XML view

Task	Command	Remarks
Enter XML view.	xml	Available in user view. To configure NETCONF in XML view, copy and paste a NETCONF message to ensure the format correctness of the NETCONF message. Do not enter the message manually. While the device is performing a NETCONF operation, do not perform any other operations, such as pasting a NETCONF message or pressing Enter. For the device to identify NETCONF messages, you must add end mark ]]>]]> at the end of each NETCONF message. Examples in this document do not have this end mark. Do add the end mark in actual operations.

Task

Command

Remarks

Enter XML view.

xml

Available in user view.

To configure NETCONF in XML view, copy and paste a NETCONF message to ensure the format correctness of the NETCONF message. Do not enter the message manually.

While the device is performing a NETCONF operation, do not perform any other operations, such as pasting a NETCONF message or pressing Enter.

For the device to identify NETCONF messages, you must add end mark ]]>]]> at the end of each NETCONF message. Examples in this document do not have this end mark. Do add the end mark in actual operations.

Exchanging capabilities

After you enter XML view, the client and the device exchange their capabilities before you can perform subsequent operations. The device automatically advertises its NETCONF capabilities to the client in a hello message as follows:

<?xml version="1.0" encoding="UTF-8"?><hello xmlns="urn:ietf:params:xml:ns:netconf:base:1.0"><capabilities><capability>urn:ietf:params:netconf:base:1.1</capability><capability>urn:ietf:params:netconf:writable-running</capability><capability>urn:ietf:params:netconf:capability:notification:1.0</capability><capability>urn:ietf:params:netconf:capability:validate:1.1</capability><capability>urn:ietf:params:netconf:capability:interleave:1.0</capability><capability>urn:h3c:params:netconf:capability:h3c-netconf-ext:1.0</capability></capabilities><session-id>1</session-id></hello>]]>]]>

The <capabilities> parameter represents the capabilities supported by the device.

The <session-id> parameter represents the unique ID assigned to the current session.

After receiving the hello message from the device, copy the following message to notify the device of the capabilities (user-configurable) supported by the client:

capability-set

</capability>

</capabilities>

</hello>

Use a pair of <capability> and </capability> tags to enclose each capability set.

Subscribing to event notifications

After you subscribe to event notifications, the device sends event notifications to the NETCONF client when a subscribed event takes place on the device. The notifications include the code, group, severity, start time, and description of the events. The device supports only log subscription. For information about which event notifications you can subscribe to, see the system log messages reference for the device.

Event subscription is not supported for NETCONF over SOAP sessions.

A subscription takes effect only on the current session. If the session is terminated, the subscription is automatically canceled.

You can send multiple subscription messages to subscribe to notification of multiple events.

Subscription procedure

# Copy the following message to the client to complete the subscription:

<?xml version="1.0" encoding="UTF-8"?>

<create-subscription xmlns="urn:ietf:params:xml:ns:netconf:notification:1.0">

<stream>NETCONF</stream>

<Group>group</Group>

<Severity>severity</Severity>

</event>

</filter>

<startTime>start-time</startTime>

</create-subscription>

</rpc>

The <stream> parameter represents the event stream type supported by the device. Only NETCONF is supported.

The <event> parameter represents an event to which you subscribe.

The <code> parameter represents a mnemonic symbol.

The <group> parameter represents the module name.

The <severity> parameter represents the severity level of the event.

The <start-time> parameter represents the start time of the subscription.

The <stop-time> argument represents the end time of the subscription.

After receiving the subscription request from the client, the device returns a response in the following format if the subscription is successful:

<?xml version="1.0" encoding="UTF-8"?>

<rpc-reply message-id="101" xmlns:netconf="urn:ietf:params:xml:ns:netconf:base:1.0">

<ok/>

</rpc-reply>

If the subscription fails, the device returns an error message in the following format:

<?xml version="1.0" encoding="UTF-8"?>

<rpc-reply message-id="101" xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<rpc-error>

<error-type>error-type</error-type>

<error-tag>error-tag</error-tag>

<error-severity>error-severity</error-severity>

<error-message xml:lang="en">error-message</error-message>

</rpc-error>

</rpc-reply>

For more information about error messages, see RFC 4741.

Example for subscribing to event notifications

Network requirements

Configure a client to subscribe to all events with no time limitation. After the subscription is successful, all events on the device are sent to the client before the session between the device and client is terminated.

Configuration procedure

# Enter XML view.

<Sysname> xml

# Notify the device of the NETCONF capabilities supported on the client.

urn:ietf:params:netconf:base:1.0

</capability>

</capabilities>

</hello>

# Subscribe to all events with no time limitation.

<create-subscription xmlns ="urn:ietf:params:xml:ns:netconf:notification:1.0">

<stream>NETCONF</stream>

</create-subscription>

</rpc>

Verifying the configuration

# If the client receives the following response, the subscription is successful:

<?xml version="1.0" encoding="UTF-8"?>

<rpc-reply xmlns="urn:ietf:params:xml:ns:netconf:base:1.0" message-id="101">

<ok/>

</rpc-reply>

# If fan 1 on the device encounters problems, the device sends the following text to the client that has subscribed to all events:

<?xml version="1.0" encoding="UTF-8"?>

<Code>FAN_DIRECTION_NOT_PREFERRED</Code>

<Severity>Alert</Severity>

<context>Fan 1 airflow direction is not preferred on slot 6, please check it.</context>

</event>

</notification>

# When another client (192.168.100.130) logs in to the device, the device sends a notification to the client that has subscribed to all events:

<?xml version="1.0" encoding="UTF-8"?>

<Group>SHELL</Group>

<Code>SHELL_LOGIN</Code>

<Severity>Notification</Severity>

<context>VTY logged in from 192.168.100.130.</context>

</event>

</notification>

Locking/unlocking the configuration

The device supports multiple NETCONF sessions. Multiple users can simultaneously manage and monitor the device using NETCONF. During device configuration and maintenance or network troubleshooting, a user can lock the configuration to prevent other users from changing it. After that, only the user holding the lock can change the configuration, and other users can only read the configuration.

In addition, only the user holding the lock can release the lock. After the lock is released, other users can change the current configuration or lock the configuration. If the session of the user that holds the lock is terminated, the system automatically releases the lock.

Locking the configuration

# Copy the following text to the client to lock the configuration:

<?xml version="1.0" encoding="UTF-8"?>

<lock>

</target>

</lock>

</rpc>

After receiving the lock request, the device returns a response in the following format if the <lock> operation is successful:

<?xml version="1.0" encoding="UTF-8"?>

<rpc-reply message-id="101"

xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<ok/>

</rpc-reply>

Unlocking the configuration

# Copy the following text to the client to unlock the configuration:

<?xml version="1.0" encoding="UTF-8"?>

</target>

</unlock>

</rpc>

After receiving the unlock request, the device returns a response in the following format if the <unlock> operation is successful:

<?xml version="1.0" encoding="UTF-8"?>

<rpc-reply message-id="101"

xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<ok/>

</rpc-reply>

Example for locking the configuration

Network requirements

Lock the device configuration so that other users cannot change the device configuration.

Configuration procedure

# Enter XML view.

<Sysname> xml

# Notify the device of the NETCONF capabilities supported on the client.

urn:ietf:params:netconf:base:1.0

</capability>

</capabilities>

</hello>

# Lock the configuration.

<?xml version="1.0" encoding="UTF-8"?>

<lock>

</target>

</lock>

</rpc>

Verifying the configuration

If the client receives the following response, the <lock> operation is successful:

<?xml version="1.0" encoding="UTF-8"?>

<rpc-reply message-id="101"

xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<ok/>

</rpc-reply>

If another client sends a lock request, the device returns the following response:

<?xml version="1.0" encoding="UTF-8"?>

<rpc-reply message-id="101" xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<rpc-error>

<error-type>protocol</error-type>

<error-tag>lock-denied</error-tag>

<error-severity>error</error-severity>

<error-message xml:lang="en"> Lock failed because the NETCONF lock is held by another session.</error-message>

<error-info>

<session-id>1</session-id>

</error-info>

</rpc-error>

</rpc-reply>

The output shows that the <lock> operation failed because the client with session ID 1 held the lock, and only the client holding the lock can release the lock.

Performing service operations

You can use NETCONF to perform service operations on the device, such as retrieving and modifying the specified information. The basic operations include get, get-bulk, get-config, get-bulk-config, and edit-config, which are used to retrieve all data, retrieve configuration data, and edit the data of the specified module. For more information, see the NETCONF XML API reference for the device.

NOTE:

During a <get>, <get-bulk>, <get-config>, or <get-bulk-config> operation, NETCONF replaces unidentifiable characters in the retrieved data with question marks (?) before sending the data to the client.

Performing the <get>/<get-bulk> operation

The <get> operation is used to retrieve device configuration and state information that match the conditions. In some cases, this operation leads to inefficiency.

The <get-bulk> operation is used to retrieve a number of data entries starting from the data entry next to the one with the specified index. One data entry contains a device configuration entry and a state information entry. The data entry quantity is defined by the count attribute, and the index is specified by the index attribute. The returned output does not include the index information. If you do not specify the index attribute, the index value starts with 1 by default.

The <get-bulk> operation retrieves all the rest data entries starting from the data entry next to the one with the specified index if either of the following conditions occurs:

· You do not specify the count attribute.

· The number of matching data entries is less than the value of the count attribute.

# Copy the following text to the client to perform the <get> operation:

<?xml version="1.0" encoding="UTF-8"?>

Specify the module, submodule, table name, and column name

</top>

</filter>

</getoperation>

</rpc>

The <getoperation> parameter can be <get> or <get-bulk>. The <filter> element is used to filter data, and it can contain module name, submodule name, table name, and column name.

· If the module name and the submodule name are not provided, the operation retrieves the data for all modules and submodules. If a module name or a submodule name is provided, the operation retrieves the data for the specified module or submodule.

· If the table name is not provided, the operation retrieves the data for all tables. If a table name is provided, the operation retrieves the data for the specified table.

· If only the index column is provided, the operation retrieves the data for all columns. If the index column and other columns are provided, the operation retrieves the data for the index column and the specified columns.

The <get> and <get-bulk> messages are similar. A <get-bulk> message carries the count and index attributes. The following is a <get-bulk> message example:

<?xml version="1.0" encoding="UTF-8"?>

<get-bulk>

<Log>

</Log>

</Logs>

</Syslog>

</top>

</filter>

</get-bulk>

</rpc>

The count attribute complies with the following rules:

· The count attribute can be placed in the module node and table node. In other nodes, it cannot be resolved.

· When the count attribute is placed in the module node, a descendant node inherits this count attribute if the descendant node does not contain the count attribute.

Verifying the configuration

After receiving the get-bulk request, the device returns a response in the following format if the operation is successful:

<?xml version="1.0"?>

<rpc-reply message-id="100"

xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<data>

Device state and configuration data

</data>

</rpc-reply>

Performing the <get-config>/<get-bulk-config> operation

The <get-config> and <get-bulk-config> operations are used to retrieve all non-default settings, which are configured by means of CLI, MIB, and Web. The <get-config> and <get-bulk-config> messages can contain the <filter> element for filtering data.

The <get-config> and <get-bulk-config> messages are similar. The following is a <get-config> message example:

<?xml version="1.0"?>

<get-config>

</source>

Specify the module name, submodule name, table name, and column name

</top>

</filter>

</get-config>

</rpc>

Verifying the configuration

After receiving the get-config request, the device returns a response in the following format if the operation is successful:

<?xml version="1.0"?>

<rpc-reply message-id="100" xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<data>

All data matching the specified filter

</data>

</rpc-reply>

Performing the <edit-config> operation

The <edit-config> operation includes the following operations: merge, create, replace, remove, delete, default-operation, error-option, test-option, and incremental. For more information about the operations, see "Supported NETCONF operations."

# Copy the following text to perform the <edit-config> operation:

<?xml version="1.0"?>

<edit-config>

<error-option>

Default operation when an error occurs

</error-option>

Specify the module name, submodule name, table name, and column name

</top>

</config>

</edit-config>

</rpc>

After receiving the edit-config request, the device returns a response in the following format if the operation is successful:

<?xml version="1.0">

<rpc-reply message-id="100" xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<ok/>

</rpc-reply>

# Perform the <get> operation to verify that the current value of the parameter is the same as the value specified through the <edit-config> operation.

All-module configuration data retrieval example

Network requirements

Retrieve configuration data for all modules.

Configuration procedure

# Enter XML view.

<Sysname> xml

# Notify the device of the NETCONF capabilities supported on the client.

urn:ietf:params:netconf:base:1.0

</capability>

</capabilities>

</hello>

# Retrieve configuration data for all modules.

<rpc message-id="100"

xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<get-config>

</source>

</get-config>

</rpc>

Verifying the configuration

If the client receives the following text, the <get-config> operation is successful:

<rpc-reply xmlns="urn:ietf:params:xml:ns:netconf:base:1.0" xmlns:web="urn:ietf:params:xml:ns:netconf:base:1.0" message-id="101">

<data>

<Ifmgr>

</Interface>

</Interface>

</Interface>

</Interface>

</Interface>

</Interfaces>

</Ifmgr>

</LogBuffer>

</Syslog>

<ZoneName>beijing</ZoneName>

</TimeZone>

</Device>

</System>

<WebUI>

</WebUI>

</Fundamentals>

</top>

</data>

</rpc-reply>

Syslog configuration data retrieval example

Network requirements

Retrieve configuration data for the Syslog module.

Configuration procedure

# Enter XML view.

<Sysname> xml

# Notify the device of the NETCONF capabilities supported on the client.

urn:ietf:params:netconf:base:1.0

</capability>

</capabilities>

</hello>

# Retrieve configuration data for the Syslog module.

<rpc message-id="100"

xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<get-config>

</source>

</top>

</filter>

</get-config>

</rpc>

Verifying the configuration

If the client receives the following text, the <get-config> operation is successful:

<?xml version="1.0" encoding="UTF-8"?>

<rpc-reply xmlns="urn:ietf:params:xml:ns:netconf:base:1.0" message-id="101">

<data>

</LogBuffer>

</Syslog>

</top>

</data>

</rpc-reply>

Example for retrieving a data entry for the interface table

Network requirements

Retrieve a data entry for the interface table.

Configuration procedure

# Enter XML view.

<Sysname> xml

# Notify the device of the NETCONF capabilities supported on the client.

<capability>urn:ietf:params:netconf:base:1.0</capability>

</capabilities>

</hello>

# Retrieve a data entry for the interface table.

<get-bulk>

<Ifmgr>

</Interfaces>

</Ifmgr>

</top>

</filter>

</get-bulk>

</rpc>

Verifying the configuration

If the client receives the following text, the <get-bulk operation> is successful:

<rpc-reply xmlns="urn:ietf:params:xml:ns:netconf:base:1.0" xmlns:web="urn:ietf:params:xml:ns:netconf:base:1.0" message-id="101">

<data>

<Ifmgr>

<Name>GigabitEthernet1/0/2</Name>

<Description>GigabitEthernet1/0/2 Interface</Description>

</Interface>

</Interfaces>

</Ifmgr>

</top>

</data>

</rpc-reply>

Example for changing the value of a parameter

Network requirements

Change the log buffer size for the Syslog module to 512.

Configuration procedure

# Enter XML view.

<Sysname> xml

# Notify the device of the NETCONF capabilities supported on the client.

<capability>urn:ietf:params:netconf:base:1.0</capability>

</capabilities>

</hello>

# Change the log buffer size for the Syslog module to 512.

<edit-config>

</target>

</LogBuffer>

</Syslog>

</top>

</config>

</edit-config>

</rpc>

Verifying the configuration

If the client receives the following text, the <edit-config> operation is successful:

<?xml version="1.0" encoding="UTF-8"?>

<rpc-reply message-id="100" xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<ok/>

</rpc-reply>

Saving, rolling back, and loading the configuration

Use NETCONF to save, roll back, or load the configuration.

Performing the <save>, <rollback>, or <load> operation consumes a lot of system resources. Do not perform these operations when the system resources are heavily occupied.

Saving the configuration

# Copy the following text to the client to save the device configuration to the specified file:

<?xml version="1.0" encoding="UTF-8"?>

<file>Specify the configuration file name</file>

</save>

</rpc>

The name of the specified configuration file must start with the storage media name and end with the .cfg extension. If the text includes the file column, you must specify the file name. The specified file will be used as the next-startup configuration file. If the text does not include the file column, the configuration is automatically saved to the default main next-startup configuration file.

The OverWrite attribute takes effect only when the name of the specified configuration file already exists. If the attribute uses the default value true, the current configuration is saved and overwrites the original file. If the attribute value is set to false, the current configuration cannot be saved, and the system displays an error message.

After receiving the save request, the device returns a response in the following format if the <save> operation is successful:

<?xml version="1.0" encoding="UTF-8"?>

<rpc-reply message-id="101"

xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<ok/>

</rpc-reply>

Rolling back the configuration based on a configuration file

# Copy the following text to the client to roll back the configuration based on a configuration file:

<?xml version="1.0" encoding="UTF-8"?>

<file>Specify the configuration file name</file>

</rollback>

</rpc>

After receiving the rollback request, the device returns a response in the following format if the <rollback> operation is successful:

<?xml version="1.0" encoding="UTF-8"?>

<rpc-reply message-id="101"

xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<ok/>

</rpc-reply>

Rolling back the configuration based on a rollback point

You can roll back the running configuration based on a rollback point when one of the following situations occurs:

· A NETCONF client sends a rollback request.

· The NETCONF session idle time is longer than the rollback idle timeout time.

· A NETCONF client is unexpectedly disconnected from the device.

To roll back the configuration based on a rollback point, perform the following tasks:

1. Lock the system.

Multiple users might simultaneously use NETCONF to configure the device. As a best practice, lock the system before rolling back the configuration to prevent other users from modifying the running configuration.

2. Mark the beginning of a <rollback> operation. For more information, see "Performing the <save-point/begin> operation."

3. Edit the device configuration. For more information, see "Performing the <edit-config> operation."

4. Configure the rollback point. For more information, see "Performing the <save-point/commit> operation."

You can repeat this step to configure multiple rollback points.

5. Roll back the configuration based on the rollback point. For more information, see "Performing the <save-point/rollback> operation."

The configuration can also be automatically rolled back based on the most recently configured rollback point when the NETCONF session idle time is longer than the rollback idle timeout time.

6. End the rollback configuration. For more information, see "Performing the <save-point/end> operation."

7. Release the lock.

Performing the <save-point/begin> operation

# Copy the following text to the client to mark the beginning of a <rollback> operation based on a rollback point:

<save-point>

<begin>

<confirm-timeout>100</confirm-timeout>

</begin>

</save-point>

</rpc>

The <confirm-timeout> parameter specifies the rollback idle timeout time in the range of 1 to 65535 seconds (the default is 600 seconds). This parameter is optional.

After receiving the begin request, the device returns a response in the following format if the <begin> operation is successful:

<rpc-reply message-id="101" xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<data>

<save-point>

<commit-id>1</commit-id>

</commit>

</save-point>

</data>

</rpc-reply>

Performing the <save-point/commit> operation

The system supports a maximum of 50 rollback points. When the limit is reached, you must specify the force attribute to overwrite the earliest rollback point.

# Copy the following text to the client to configure the rollback point:

<save-point>

<label>SUPPORT VLAN<label>

<comment>vlan 1 to 100 and interfaces. Each vlan used for different custom as fllows: ……</comment>

</commit>

</save-point>

</rpc>

The <label> and <comment> parameters are optional.

After receiving the commit request, the device returns a response in the following format if the <commit> operation is successful:

<rpc-reply message-id="101" xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<data>

<save-point>

<commit-id>2</commit-id>

</commit>

</save-point>

</data>

</rpc-reply>

Performing the <save-point/rollback> operation

# Copy the following text to the client to roll back the configuration:

<save-point>

<commit-id/>

<commit-index/>

<commit-label/>

</rollback>

</save-point>

</rpc>

The <commit-id> parameter uniquely identifies a rollback point.

The <commit-index> parameter specifies 50 most recently configured rollback points. The value of 0 indicates the most recently configured one and 49 indicates the earliest configured one.

The <commit-label> parameter exclusively specifies a label for a rollback point. The label is not required for a rollback point.

Specify one of these parameters to roll back the specified configuration. If no parameter is specified, this operation rolls back configuration based on the most recently configured rollback point.

After receiving the rollback request, the device returns a response in the following format if the <rollback> operation is successful:

<rpc-reply message-id="101" xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

</rpc-reply>

Performing the <save-point/end> operation

# Copy the following text to the client to end the rollback configuration:

<save-point>

<end/>

</save-point>

</rpc>

After receiving the end request, the device returns a response in the following format if the <end> operation is successful:

<rpc-reply message-id="100" xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<ok/>

</rpc-reply>

Performing the <save-point/get-commits> operation

# Copy the following text to the client to get the rollback point configuration records:

<save-point>

<get-commits>

<commit-id/>

<commit-index/>

<commit-label/>

</get-commits>

</save-point>

</rpc>

Specify one of the <commit-id>, <commit-index>, and <commit-label> parameters to get the specified rollback point configuration records. If no parameter is specified, this operation gets records for all rollback point settings. The following text is a <save-point>/<get-commits> request example:

<save-point>

<get-commits>

<commit-label>SUPPORT VLAN</commit-label>

</get-commits>

</save-point>

</rpc>

After receiving the get commits request, the device returns a response in the following format if the <get-commits> operation is successful:

<rpc-reply message-id="101" xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<data>

<save-point>

<commit-information>

<Label>SUPPORT VLAN</Label>

</commit-information>

</save-point>

</data>

</rpc-reply>

Performing the <save-point/get-commit-information> operation

# Copy the following text to the client to get the system configuration data corresponding to a rollback point:

<save-point>

<get-commit-information>

<commit-information>

<commit-id/>

<commit-index/>

<commit-label/>

</commit-information>

<compare-information>

<commit-id/>

<commit-index/>

<commit-label/>

</compare-information

</get-commit-information>

</save-point>

</rpc>

Specify one of the <commit-id>, <commit-index>, and <commit-label> parameters to get the configuration data corresponding to the specified rollback point. The <compare-information> parameter is optional. If no parameter is specified, this operation gets the configuration data corresponding to the most recently configured rollback point. The following text is a <save-point>/<get-commit-information> request example:

<save-point>

<get-commit-information>

<commit-information>

<commit-label>SUPPORT VLAN</commit-label>

</commit-information>

</get-commit-information>

</save-point>

</rpc>

After receiving the get-commit-information request, the device returns a response in the following format if the <get-commit-information> operation is successful:

<rpc-reply message-id="101" xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<data>

<save-point>

<commit-information>

…

interface vlan 1

…

</content>

</commit-information>

</save-point>

</data>

</rpc-reply>

Loading the configuration

After you perform a <load> operation, the loaded settings are merged into the current configuration as follows:

· New settings are directly loaded.

· Settings that already exist in the current configuration are replaced by those loaded from the configuration file.

Some settings in a configuration file might conflict with the existing settings. For the settings in the file to take effect, delete the existing conflicting settings, and then load the configuration file.

# Copy the following text to the client to load a configuration file for the device:

<?xml version="1.0" encoding="UTF-8"?>

<load>

<file>Specify the configuration file name</file>

</load>

</rpc>

The name of the specified configuration file must start with the storage media name and end with the .cfg extension.

After receiving the load request, the device returns a response in the following format if the <load> operation is successful:

<?xml version="1.0" encoding="UTF-8"?>

<rpc-reply message-id="101"

xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<ok/>

</rpc-reply>

Example for saving the configuration

Network requirements

Save the current configuration to configuration file my_config.cfg.

Configuration procedure

# Enter XML view.

<Sysname> xml

# Notify the device of the NETCONF capabilities supported on the client.

urn:ietf:params:netconf:base:1.0

</capability>

</capabilities>

</hello>

# Save the configuration of the device to configuration file my_config.cfg.

<?xml version="1.0" encoding="UTF-8"?>

<save>

<file>my_config.cfg</file>

</save>

</rpc>

Verifying the configuration

If the client receives the following response, the <save> operation is successful:

<?xml version="1.0" encoding="UTF-8"?>

<rpc-reply message-id="101"

xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<ok/>

</rpc-reply>

Filtering data

You can define a filter to filter information when you perform a <get>, <get-bulk>, <get-config>, or <get-bulk-config> operation. Data filtering includes the following types:

· Table-based filtering—Filters table information.

· Column-based filtering—Filters information for a single column.

For table-based filtering to take effect, you must configure table-based filtering before column-based filtering.

Table-based filtering

You can specify a match criterion for the row attribute filter to implement table-based filtering, for example, IP address filtering. The namespace is http://www.h3c.com/netconf/base:1.0. For information about the support for table-based match, see NETCONF XML API documents.

# Copy the following text to the client to retrieve the longest data with VRF name vpn1, IP address 1.1.1.0, and mask length 24 from the IPv4 routing table:

<get>

<Route>

</Ipv4Routes>

</Route>

</top>

</filter>

</get>

</rpc>

Column-based filtering

Column-based filtering includes full match filtering, regular expression match filtering, and conditional match filtering. Full match filtering has the highest priority and conditional match filtering has the lowest priority. When more than one filtering criterion is specified, the one with the highest priority takes effect.

Full match filtering

You can specify an element value in an XML message to implement full match filtering. If multiple element values are provided, the system returns the data that matches all the specified values.

# Copy the following text to the client to retrieve configuration data of all interfaces in UP state:

<get>

<Ifmgr>

</Interface>

</Interfaces>

</Ifmgr>

</top>

</filter>

</get>

</rpc>

You can also specify an attribute name that is the same as a column name of the current table at the row to implement full match filtering. The system returns only configuration data that matches this attribute name. The XML message equivalent to the above element-value-based full match filtering is as follows:

<get>

<Ifmgr>

</Interfaces>

</Ifmgr>

</top>

</filter>

</get>

</rpc>

The above examples show that both element-value-based full match filtering and attribute-name-based full match filtering can retrieve the same configuration data for all UP interfaces.

Regular expression match filtering

To implement a complex data filtering with characters, you can add a regExp attribute for a specific element.

The supported data types include integer, date and time, character string, IPv4 address, IPv4 mask, IPv6 address, MAC address, OID, and time zone.

# Copy the following text to the client to retrieve the descriptions of interfaces, of which all the characters must be upper-case letters from A to Z:

<get-config>

</source>

<Ifmgr>

</Interface>

</Interfaces>

</Ifmgr>

</top>

</filter>

</get-config>

</rpc>

Conditional match filtering

To implement a complex data filtering with digits and character strings, you can add a match attribute for a specific element. Table 3 lists the conditional match operators.

Table 3 Conditional match operators

Operation	Operator	Remarks
More than	match="more:value"	More than the specified value. The supported data types include date, digit, and character string.
Less than	match="less:value"	Less than the specified value. The supported data types include date, digit, and character string.
Not less than	match="notLess:value"	Not less than the specified value. The supported data types include date, digit, and character string.
Not more than	match="notMore:value"	Not more than the specified value. The supported data types include date, digit, and character string.
Equal	match="equal:value"	Equal to the specified value. The supported data types include date, digit, character string, OID, and BOOL.
Not equal	match="notEqual:value"	Not equal to the specified value. The supported data types include date, digit, character string, OID, and BOOL.
Include	match="include:string"	Includes the specified string. The supported data types include only character string.
Not include	match="exclude:string"	Excludes the specified string. The supported data types include only character string.
Start with	match="startWith:string"	Starts with the specified string. The supported data types include character string and OID.
End with	match="endWith:string"	Ends with the specified string. The supported data types include only character string.

# Copy the following text to the client to retrieve extension information about the entity whose CPU usage is more than 50%:

<get>

</Entity>

</ExtPhysicalEntities>

</Device>

</top>

</filter>

</get>

</rpc>

Example for filtering data with regular expression match

Network requirements

Retrieve all data including Gigabit in the Description column of the Interfaces table under the Ifmgr module.

Configuration procedure

# Enter XML view.

<Sysname> xml

# Notify the device of the NETCONF capabilities supported on the client.

urn:ietf:params:netconf:base:1.0

</capability>

</capabilities>

</hello>

# Retrieve all data including Gigabit in the Description column of the Interfaces table under the Ifmgr module.

<?xml version="1.0"?>

<rpc message-id="100"

xmlns="urn:ietf:params:xml:ns:netconf:base:1.0" xmlns:reg="http://www.h3c.com/netconf/base:1.0">

<get>

<Ifmgr>

</Interface>

</Interfaces>

</Ifmgr>

</top>

</filter>

</get>

</rpc>

Verifying the configuration

If the client receives the following text, the operation is successful:

<?xml version="1.0" encoding="UTF-8"?>

<rpc-reply xmlns="urn:ietf:params:xml:ns:netconf:base:1.0" xmlns:reg="http://www.h3c.com/netconf/base:1.0" message-id="100">

<data>

<Ifmgr>

<Description>GigabitEthernet1/0/1 Interface</Description>

</Interface>

<Description>GigabitEthernet1/0/2 Interface</Description>

</Interface>

<Description>GigabitEthernet1/0/3 Interface</Description>

</Interface>

</Ifmgr>

</top>

</data>

</rpc-reply>

Example for filtering data by conditional match

Network requirements

Retrieve data in the Name column with the ifindex value not less than 5000 in the Interfaces table under the Ifmgr module.

Configuration procedure

# Enter XML view.

<Sysname> xml

# Notify the device of the NETCONF capabilities supported on the client.

urn:ietf:params:netconf:base:1.0

</capability>

</capabilities>

</hello>

# Retrieve data in the Name column with the ifindex value not less than 5000 in the Interfaces table under the Ifmgr module.

<rpc message-id="100"

xmlns="urn:ietf:params:xml:ns:netconf:base:1.0" xmlns:nc="http://www.h3c.com/netconf/base:1.0">

<get>

<Ifmgr>

<Name/>

</Interface>

</Interfaces>

</Ifmgr>

</top>

</filter>

</get>

</rpc>

Verifying the configuration

If the client receives the following text, the operation is successful:

<?xml version="1.0" encoding="UTF-8"?>

<rpc-reply xmlns="urn:ietf:params:xml:ns:netconf:base:1.0" xmlns:nc="http://www.h3c.com/netconf/base:1.0" message-id="100">

<data>

<Ifmgr>

</Interface>

<Name>Register-Tunnel0</Name>

</Interface>

</Interfaces>

</Ifmgr>

</top>

</data>

</rpc-reply>

Performing CLI operations through NETCONF

You can enclose command lines in XML messages to configure the device.

Performing <CLI> operations through NETCONF is resource intensive. As a best practice, do not perform the following tasks:

· Enclose multiple command lines in one XML message.

· Use NETCONF to perform a <CLI> operation when other users are performing NETCONF CLI operations.

Configuration procedure

# Copy the following text to the client to execute the commands:

<?xml version="1.0" encoding="UTF-8"?>

<CLI>

Commands

</Execution>

</CLI>

</rpc>

The <Execution> element can contain multiple commands, with one command on one line.

After receiving the <CLI> operation request, the device returns a response in the following format if the <CLI> operation is successful:

<?xml version="1.0" encoding="UTF-8"?>

<rpc-reply message-id="101" xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<CLI>

<![CDATA[Responses to the commands]]>

</Execution>

</CLI>

</rpc-reply>

CLI operation example

Configuration requirements

Send the display current-configuration command to the device.

Configuration procedure

# Enter XML view.

<Sysname> xml

# Notify the device of the NETCONF capabilities supported on the client.

urn:ietf:params:netconf:base:1.0

</capability>

</capabilities>

</hello>

# Copy the following text to the client to execute the display current-configuration command:

<?xml version="1.0" encoding="UTF-8"?>

<CLI>

display current-configuration

</Execution>

</CLI>

</rpc>

Verifying the configuration

If the client receives the following text, the operation is successful:

<?xml version="1.0" encoding="UTF-8"?>

<rpc-reply message-id="101" xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<CLI>

<Execution><![CDATA[

<Sysname>display current-configuration

version 7.1.052, Demo 2501005

sysname Sysname

ftp server enable

ftp update fast

ftp timeout 2000

irf mac-address persistent timer

irf auto-update enable

undo irf link-delay

domain default enable system

telnet server enable

vlan 1

vlan 1000

radius scheme system

primary authentication 127.0.0.1 1645

return

]]>

</Execution>

</CLI>

</rpc-reply>

Retrieving NETCONF information

# Copy the following text to the client to retrieve NETCONF information:

<?xml version="1.0" encoding="UTF-8"?>

<get>

<netconf-state xmlns='urn:ietf:params:xml:ns:yang:ietf-netconf-monitoring'>

</netconf-state>

</filter>

</get>

</rpc>

The value for the <getType> parameter can be one of the following operations:

· capabilities—Retrieves device capabilities.

· datastores—Retrieves databases from the device.

· schemas—Retrieves the list of the YANG file names from the device.

· sessions—Retrieves session information from the device.

· statistics—Retrieves NETCONF statistics.

If you do not specify a value for the <getType> parameter, the retrieval operation retrieves all NETCONF information.

The retrieval operation does not support data filtering.

After receiving the NETCONF information retrieval request, the device returns a response in the following format if the operation is successful:

<?xml version="1.0"?>

<rpc-reply message-id="100"

xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<data>

ALL NETCONF information

</data>

</rpc-reply>

Retrieving YANG file content

YANG files save the NETCONF operations supported by the device. A user can know the supported operations by retrieving and analyzing the content of YANG files.

YANG files are integrated in the device software and are named in the format of yang_identifier@yang_version.yang. You cannot view the YANG file names by executing the dir command. For information about how to retrieve the YANG file names, see "Retrieving NETCONF information."

# Copy the following text to the client to retrieve the YANG file named [email protected]:

<get-schema xmlns='urn:ietf:params:xml:ns:yang:ietf-netconf-monitoring'>

<identifier>syslog-data</identifier>

</get-schema>

</rpc>

# After receiving the YANG file retrieve request, the device returns a response in the following format if the operation is successful:

<?xml version="1.0"?>

<rpc-reply message-id="100"

xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<data>

Content of the specified YANG file

</data>

</rpc-reply>

Retrieving NETCONF session information

You can use the <get-sessions> operation to retrieve NETCONF session information of the device.

# Copy the following message to the client to retrieve NETCONF session information from the device:

<?xml version="1.0" encoding="UTF-8"?>

<get-sessions/>

</rpc>

After receiving the get-sessions request, the device returns a response in the following format if the <get-sessions> operation is successful:

<?xml version="1.0" encoding="UTF-8"?>

<rpc-reply message-id="101" xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<get-sessions>

<SessionID>Configuration session ID</SessionID>

<Line>Line information</Line>

<UserName>Name of the user creating the session</UserName>

<Since>Time when the session was created</Since>

<LockHeld>Whether the session holds a lock</LockHeld>

</Session>

</get-sessions>

</rpc-reply>

For example, to get NETCONF session information:

# Enter XML view.

<Sysname> xml

# Notify the device of the NETCONF capabilities supported on the client.

urn:ietf:params:netconf:base:1.0

</capability>

</capabilities>

</hello>

# Copy the following message to the client to get the current NETCONF session information on the device:

<?xml version="1.0" encoding="UTF-8"?>

<get-sessions/>

</rpc>

If the client receives a message as follows, the operation is successful:

<?xml version="1.0" encoding="UTF-8"?>

<rpc-reply xmlns="urn:ietf:params:xml:ns:netconf:base:1.0" message-id="101">

<get-sessions>

<LockHeld>false</LockHeld>

</Session>

</get-sessions>

</rpc-reply>

The output shows the following information:

· The session ID of an existing NETCONF session is 1.

· The login user type is vty0.

· The login time is 2011-01-05T00:24:57.

· The user does not hold the lock of the configuration.

Terminating another NETCONF session

NETCONF allows one client to terminate the NETCONF session of another client. The client whose session is terminated returns to user view.

Configuration procedure

# Copy the following message to the client to terminate the specified NETCONF session:

<kill-session>

<session-id>

Specified session-ID

</session-id>

</kill-session>

</rpc>

After receiving the kill-session request, the device returns a response in the following format if the <kill-session> operation is successful:

<?xml version="1.0" encoding="UTF-8"?>

<rpc-reply message-id="101"

xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<ok/>

</rpc-reply>

Configuration example

Configuration requirement

The user whose session's ID is 1 terminates the session with session ID 2.

Configuration procedure

# Enter XML view.

<Sysname> xml

# Notify the device of the NETCONF capabilities supported on the client.

urn:ietf:params:netconf:base:1.0

</capability>

</capabilities>

</hello>

# Terminate the session with session ID 2.

<kill-session>

<session-id>2</session-id>

</kill-session>

</rpc>

Verifying the configuration

If the client receives the following text, the NETCONF session with session ID 2 has been terminated, and the client with session ID 2 has returned from XML view to user view:

<?xml version="1.0" encoding="UTF-8"?>

<rpc-reply message-id="101" xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<ok/>

</rpc-reply>

Returning to the CLI

To return from XML view to the CLI, send the following close-session request:

<close-session/>

</rpc>

When the device receives the close-session request, it sends the following response and returns to CLI's user view:

<?xml version="1.0" encoding="UTF-8"?>

<rpc-reply message-id="101"

xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">

<ok/>

</rpc-reply>

Supported NETCONF operations

This chapter describes NETCONF operations available with Comware 7.

action

Usage guidelines

This operation issues actions for non-default settings, for example, reset action.

XML example

# Clear statistics information for all interfaces.

<Ifmgr>

<Clear>

</Clear>

</ClearAllIfStatistics>

</Ifmgr>

</top>

</action>

</rpc>

CLI

Usage guidelines

This operation executes CLI commands.

A request message encloses commands in the <CLI> element. A response message encloses the command output in the <CLI> element.

You can use the following elements to execute commands:

· Execution—Executes commands in user view.

· Configuration —Executes commands in system view. To execute commands in a lower-level view of the system view, use the <Configuration> element to enter the view first.

To use this element, include the exec-use-channel attribute and specify a value for the attribute:

¡ false—Executes commands without using a channel.

¡ true—Executes commands by using a temporary channel. The channel is automatically closed after the execution.

¡ persist—Executes commands by using the persistent channel for the session.

To use the persistent channel, first perform an <Open-channel> operation to open the persistent channel. If you do not do so, the system will automatically open the persistent channel.

After using the persistent channel, perform a <Close-channel> operation to close the channel and return to system view. If you do not perform an <Open-channel> operation, the system stays in the view and will execute subsequent commands in the view.

You can also specify the error-when-rollback attribute in the <Configuration> element to indicate whether CLI operations are allowed during a configuration error-triggered configuration rollback. This attribute takes effect only if the value of the <error-option> element in <edit-config> operations is set to rollback-on-error. It has the following values:

¡ true—Rejects CLI operation requests and returns error messages.

¡ false (the default)—Allows CLI operations.

For CLI operations to be correctly performed, set the value of the error-when-rollback attribute to true.

A NETCONF session supports only one persistent channel and but supports multiple temporary channels.

NETCONF does not support executing interactive commands.

You cannot execute the quit command by using a channel to exit user view.

XML example

# Execute the vlan 3 command in system view without using a channel.

<CLI>

</CLI>

</rpc>

close-session

Usage guidelines

This operation terminates the current NETCONF session, unlock the configuration, and release the resources (for example, memory) used by the session. After this operation, you exit the XML view.

XML example

# Terminate the current NETCONF session.

<close-session/>

</rpc>

edit-config: create

Usage guidelines

This operation creates target configuration items.

To use the create attribute in an <edit-config> operation, you must specify the target configuration item.

· If the table supports creating a target configuration item and the item does not exist, the operation creates the item and configures the item.

· If the specified item already exists, a data-exist error message is returned.

XML example

# Set the buffer size to 120.

<edit-config>

</target>

</LogBuffer>

</Syslog>

</top>

</config>

</edit-config>

</rpc>

edit-config: delete

Usage guidelines

This operation deletes the specified configuration.

· If the specified target has only the table index, the operation removes all configuration of the specified target, and the target itself.

· If the specified target has the table index and configuration data, the operation removes the specified configuration data of this target.

· If the specified target does not exist, an error message is returned, showing that the target does not exist.

XML example

The syntax is the same as the edit-config message with the create attribute. Change the operation attribute from create to delete.

edit-config: merge

Usage guidelines

This operation commits target configuration items to the running configuration.

To use the merge attribute in an <edit-config> operation, you must specify the target configuration item (on a specific level):

· If the specified item exists, the operation directly updates the setting for the item.

· If the specified item does not exist, the operation creates the item and configures the item.

· If the specified item does not exist and it cannot be created, an error message is returned.

XML example

The XML data format is the same as the edit-config message with the create attribute. Change the operation attribute from create to merge.

edit-config: remove

Usage guidelines

This operation removes the specified configuration.

· If the specified target has only the table index, the operation removes all configuration of the specified target, and the target itself.

· If the specified target has the table index and configuration data, the operation removes the specified configuration data of this target.

· If the specified target does not exist, or the XML message does not specify any targets, a success message is returned.

XML example

The syntax is the same as the edit-config message with the create attribute. Change the operation attribute from create to remove.

edit-config: replace

Usage guidelines

This operation replaces the specified configuration.

· If the specified target exists, the operation replaces the configuration of the target with the configuration carried in the message.

· If the specified target does not exist but is allowed to be created, the operation creates the target and then applies the configuration.

· If the specified target does not exist and is not allowed to be created, the operation is not conducted and an invalid-value error message is returned.

XML example

The syntax is the same as the edit-config message with the create attribute. Change the operation attribute from create to replace.

edit-config: test-option

Usage guidelines

This operation determines whether to commit a configuration item in an <edit-configure> operation. The <test-option> element has one of the following values:

· test-then-set—Performs a syntax check, and commits an item if the item passes the check. If the item fails the check, the item is not committed. This is the default test-option value.

· set—Commits the item without performing a syntax check.

· test-only—Performs only a syntax check. If the item passes the check, a success message is returned. Otherwise, an error message is returned.

XML example

# Test the configuration for an interface.

<edit-config>

</target>

<test-option>test-only</test-option>

</Interface>

</Interfaces>

</Ifmgr>

</top>

</config>

</edit-config>

</rpc>

edit-config: default-operation

Usage guidelines

This operation modifies the running configuration of the device by using the default operation method.

NETCONF uses one of the following operation attributes to modify configuration: merge, create, delete, and replace If you do not specify an operation attribute for an edit-config message, NETCONF uses the default operation method. Your setting of the value for the <default-operation> element takes effect only once. If you do not specify an operation attribute or the default operation method for an <edit-config> message, merge always applies.

The <default-operation> element has the following values:

· merge—Default value for the <default-operation> element.

· replace—Value used when the operation attribute is not specified and the default operation method is specified as replace.

· none—Value used when the operation attribute is not specified and the default operation method is specified as none. If this value is specified, the <edit-config> operation is used only for schema verification rather than issuing a configuration. If the schema verification is passed, a successful message is returned. Otherwise, an error message is returned.

XML example

# Issue an empty operation for schema verification purposes.

<edit-config>

</target>

<default-operation>none</default-operation>

<Ifmgr>

</Interface>

</Interfaces>

</Ifmgr>

</top>

</config>

</edit-config>

</rpc>

edit-config: error-option

Usage guidelines

This operation determines the action to take in case of a configuration error.

The <error-option> element has the following values:

· stop-on-error—Stops the operation and returns an error message. This is the default error-option value.

· continue-on-error—Continues the operation and returns an error message.

· rollback-on-error—Rolls back the configuration.

XML example

# Issue the configuration for two interfaces with the <error-option> element value as continue-on-error.

<edit-config>

</target> <error-option>continue-on-error</error-option>

</Interface>

</Interface>

</Interfaces>

</Ifmgr>

</top>

</config>

</edit-config>

</rpc>

edit-config: incremental

Usage guidelines

This operation adds configuration data to a column without affecting the original data.

The incremental attribute applies to a list column such as the vlan permitlist column.

You can use the incremental attribute for <edit-config> operations except the <replace> operation.

Support for the incremental attribute varies by module. For more information, see NETCONF XML API documents for the switch.

XML example

# Add VLANs 1 through 10 to an untagged VLAN list that has untagged VLANs 12 through 15.

<rpc message-id="100" xmlns="urn:ietf:params:xml:ns:netconf:base:1.0"

xmlns:h3c="http://www.h3c.com/netconf/base:1.0">

<edit-config>

</target>

</Interface>

</HybridInterfaces>

</VLAN>

</top>

</config>

</edit-config>

</rpc>

get

Usage guidelines

This operation retrieves device configuration and state information.

XML example

# Retrieve device configuration and state information for the Syslog module.

<get>

</Syslog>

</top>

</filter>

</get>

</rpc>

get-bulk

Usage guidelines

This operation retrieves a number of data entries (including device configuration and state information) starting from the data entry next to the one with the specified index.

XML example

# Retrieve device configuration and state information for all interfaces.

<get-bulk>

<Ifmgr>

</Interfaces>

</Ifmgr>

</top>

</filter>

</get-bulk>

</rpc>

get-bulk-config

Usage guidelines

This operation retrieves a number of non-default configuration data entries starting from the data entry next to the one with the specified index.

XML example

# Retrieve non-default configuration for all interfaces.

<get-bulk-config>

</source>

<Ifmgr>

</Ifmgr>

</top>

</filter>

</get-bulk-config>

</rpc>

get-config

Usage guidelines

This operation retrieves non-default configuration data. If no non-default configuration data exists, the device returns a response with empty data.

XML example

# Retrieve non-default configuration data for the interface table.

<get-config>

</source>

<Ifmgr>

</Interfaces>

</Ifmgr>

</top>

</filter>

</get-config>

</rpc>

get-sessions

Usage guidelines

This operation retrieves information about all NETCONF sessions in the system. You cannot specify a session ID to retrieve information about a specific NETCONF session.

XML example

# Retrieve information about all NETCONF sessions in the system.

<get-sessions/>

</rpc>

kill-session

Usage guidelines

This operation terminates the NETCONF session for another user. This operation cannot terminate the NETCONF session for the current user.

XML example

# Terminate the NETCONF session with session ID 1.

<kill-session>

<session-id>1</session-id>

</kill-session>

</rpc>

load

Usage guidelines

This operation loads the configuration. After the device finishes a <load> operation, the configuration in the specified file is merged into the running configuration of the device.

XML example

# Merge the configuration in file a1.cfg to the running configuration of the device.

<load>

</load>

</rpc>

lock

Usage guidelines

This operation locks the configuration. After the configuration is locked, you cannot perform <edit-config> operations. Other operations are allowed.

After a user locks the configuration, other users cannot use NETCONF or any other configuration methods such as CLI and SNMP to configure the device.

XML example

# Lock the configuration.

<lock>

</target>

</lock>

</rpc>

rollback

Usage guidelines

This operation rolls back the configuration. To do so, you must specify the configuration file in the <file> element. After the device finishes the <rollback> operation, the current device configuration is totally replaced with the configuration in the specified configuration file.

XML example

# Roll back the running configuration to the configuration in file 1A.cfg.

</rollback>

</rpc>

save

Usage guidelines

This operation saves the running configuration. You can use the <file> element to specify a file for saving the configuration. If the text does not include the file column, the running configuration is automatically saved to the main next-startup configuration file.

The OverWrite attribute determines whether the running configuration overwrites the original configuration file when the specified file already exists.

The Binary-only attribute determines whether to save the running configuration only to the binary configuration file.

XML example

# Save the running configuration to file test.cfg.

</save>

</rpc>

unlock

Usage guidelines

This operation unlocks the configuration, so other users can configure the device.

Terminating a NETCONF session automatically unlocks the configuration.

XML example

# Unlock the configuration.

</target>

</unlock>

</rpc>

12-Network Management and Monitoring Configuration Guide

NETCONF message format

NETCONF

NETCONF over SOAP

Exchanging capabilities

Example for subscribing to event notifications

Network requirements

Configuration procedure

Verifying the configuration

Network requirements

Configuration procedure

Verifying the configuration

Verifying the configuration

Verifying the configuration

All-module configuration data retrieval example

Network requirements

Configuration procedure

Verifying the configuration

Syslog configuration data retrieval example

Network requirements

Configuration procedure

Verifying the configuration

Network requirements

Configuration procedure

Verifying the configuration

Network requirements

Configuration procedure

Verifying the configuration

Rolling back the configuration based on a configuration file

Rolling back the configuration based on a rollback point

Performing the <save-point/get-commits> operation

Performing the <save-point/get-commit-information> operation

Network requirements

Configuration procedure

Verifying the configuration

Full match filtering

Regular expression match filtering

Conditional match filtering

Network requirements

Configuration procedure

Verifying the configuration

Network requirements

Configuration procedure

Verifying the configuration

Configuration requirements

Configuration procedure

Verifying the configuration

Configuration requirement

Configuration procedure

Verifying the configuration

edit-config: test-option

Intelligent Terminal Products

Product Support Services

Technical Service Solutions

Resource Center

Policy

Online Help

Become a Partner

Partner Policy & Program

Global Learning

Partner Sales Resources

Service Business

News & Events