3.7. Network QoS Policy Command Reference

This command creates a text description stored in the configuration file for a configuration context.

The description command associates a text string with a configuration context to help identify the context in the configuration file.

The no form of this command removes any description string from the context.

This command creates a profile for the user to configure the ingress QoS parameters of a Multi-Link Frame Relay (MLFR) bundle. A maximum of 128 ingress QoS profiles may be created on the system.

The no form of this command deletes the profile.

This command provides the Frame Relay scheduling class context for the user to configure the ingress or egress QoS parameters of an MLFR bundle or an FRF.12 UNI/NNI link for this profile.

This command configures the value of the MLFR bundle ingress per-class reassembly timer for the profile.

This command creates a profile for the user to configure the egress QoS parameters of an MLFR bundle or an FRF.12 UNI/NNI link. A maximum of 128 egress QoS profiles may be created on the system.

The no form of this command deletes the profile.

This command configures the maximum size for each Frame Relay scheduling class queue for this profile.

This command configures the minimum information rate scheduling parameter for each Frame Relay scheduling class queue for this profile.

This command configures the WRR weight scheduling parameter for each Frame Relay scheduling class queue for this profile.

This command creates or edits a QoS network policy. The network policy defines the treatment that IP or MPLS packets receive as they ingress and egress the network port.

The QoS network policy consists of an ingress and egress component. The ingress component of the policy defines how DiffServ code points and MPLS EXP bits are mapped to internal forwarding class and profile state. The forwarding class and profile state define the Per Hop Behavior (PHB) or the QoS treatment through the router. The mapping on each network interface defaults to the mappings defined in the default network QoS policy until an explicit policy is defined for the network interface.

The egress component of the network QoS policy defines the queuing parameters associated with each forwarding class. Each of the forwarding classes defined within the system automatically creates a queue on each network interface. This queue gets all the parameters defined within the default network QoS policy 1 until an explicit policy is defined for the network interface access uplink port. If the egressing packet originated on an ingress SAP, or the remarking parameter is defined for the egress interface, the egress QoS policy also defines the IP DSCP, dot1p/DE, or MPLS EXP bit marking based on the forwarding class and the profile state.

Network policy-id 1 exists as the default policy that is applied to all network interfaces by default. The network policy-id 1 cannot be modified or deleted. It defines the default DSCP-to-FC mapping and MPLS EXP-to-FC mapping for the ingress. For the egress, it defines six forwarding classes that represent individual queues and the packet marking criteria.

Network policy-id 1 exists as the default policy that is applied to all network ports by default. This default policy cannot be modified or deleted. It defines the default DSCP-to-FC mapping and default unicast meters for ingress IP traffic. For the egress, it defines the forwarding class to dot1p and DSCP values and the packet marking criteria.

If a new network policy is created (for instance, policy-id 3), only the default action and egress forwarding class parameters are identical to the default policy. A new network policy does not contain the default DSCP-to-FC and MPLS-EXP-to-FC mapping for network QoS policy of type ip-interface or the DSCP-to-FC mapping (for network QoS policy of type port). The default network policy can be copied (use the copy command) to create a new network policy that includes the default ingress DSCP-to-FC and MPLS EXP-to-FC mapping (as appropriate). You can modify parameters or use the no modifier to remove an object from the configuration.

Any changes made to an existing policy, using any of the sub-commands, will be applied immediately to all network interfaces where this policy is applied. For this reason, when many changes are required on a policy, it is highly recommended that the policy be copied to a work area policy-id. That work-in-progress policy can be modified until complete, then written over the original policy-id. Use the config qos copy command to maintain policies in this manner.

The no form of this command deletes the network policy. A policy cannot be deleted until it is removed from all entities where it is applied. The default network policy policy-id 1 cannot be deleted.

This command configures the network policy scope as exclusive or template. The policy’s scope cannot be changed if the policy is applied to an interface.

The no form of this command sets the scope of the policy to the default of template.

This command is used to enter the CLI node that creates or edits policy entries that specify the DiffServ code points-to-forwarding class mapping for all IP packets and define the MPLS EXP bits-to-forwarding class mapping for all labeled packets.

When premarked IP or MPLS packets ingress on a network port, they get a Per Hop Behavior (that is, the QoS treatment through the router, based on the mapping defined under the current node).

This command defines or edits the default action to be taken for packets that have an undefined DSCP or MPLS EXP bit set. The default-action command specifies the forwarding class to which such packets are assigned.

Multiple default-action commands will overwrite each previous default-action command.

This command explicitly sets the forwarding class or enqueuing priority and profile of the packet when a packet is marked with a dot1p-priority specified. Adding a dot1p rule on the policy forces packets that match the dot1p-priority specified to override and be assigned to the forwarding class and enqueuing priority and profile of the packet, based on the parameters included in the dot1p rule. When the forwarding class is not specified in the rule, a matching packet preserves (or inherits) the existing forwarding class derived from earlier matches in the classification hierarchy. When the enqueuing priority is not specified in the rule, a matching packet preserves (or inherits) the existing enqueuing priority derived from earlier matches in the classification hierarchy.

The dot1p-priority is derived from the most significant three bits in the IEEE 802.1q or IEEE 802.1p header. The three dot1p bits define 8 Class-of-Service (CoS) values commonly used to map packets to per-hop Quality of Service (QoS) behavior.

The no form of this command removes the explicit dot1p classification rule from the policy. Removing the rule on the policy immediately removes the rule on all ingress SAPs using the policy.

This command creates a mapping between the DiffServ Code Point (DSCP) of the network ingress traffic and the forwarding class.

Ingress traffic that matches the specified DSCP will be assigned to the corresponding forwarding class. Multiple commands can be entered to define the association of some or all 64 DiffServ code points to the forwarding class. For undefined code points, packets are assigned to the forwarding class specified under the default-action command.

The no form of this command removes the DiffServ code point-to-forwarding class association. The default-action then applies to that code point value.

This command specifies the forwarding class name. The forwarding class name represents an egress queue. The fc fc-name represents a CLI parent node that contains sub-commands or parameters describing the egress characteristics of the queue and the marking criteria of packets flowing through it. The fc command overrides the default parameters for that forwarding class to the values defined in the network default policy.

The no form of this command removes the forwarding class name associated with this queue, as appropriate. The forwarding class reverts to the defined parameters in the default network policy. If the fc-name is removed from the network policy, that forwarding class reverts to the factory defaults.

This command is used to redirect the FC of a broadcast packet received in a VPLS service over a PW or network IP interface to an ingress forwarding plane queue-group.

It defines the mapping of an FC to a policer-id and redirects the lookup of the policer of the same ID in some ingress forwarding plane queue-group instance. However, the queue-group name and instance are explicitly provided only at the time the network QoS policy is applied to the ingress context of a spoke or mesh SDP or a network IP interface.

The broadcast-policer statement is ignored when the network QoS policy is applied to any object other than a VPLS spoke or mesh SDP or a network IP interface.

The no form of this command removes the redirection of the FC.

IP criteria-based network ingress policies are used to select the appropriate ingress queue and corresponding forwarding class for matched traffic. This command is used to enter the context to create or edit policy entries that specify IP criteria such as IP quintuple lookup or DiffServ code point.

The 7750 SR OS implementation will exit on the first match found and execute the actions in accordance with the accompanying action command. For this reason, entries must be sequenced correctly from most to least explicit.

The classification only applies to the outer IP header of non-tunneled traffic. The only exception is for traffic received on a Draft Rosen tunnel, for which classification on the outer IP header only is supported.

Attempting to apply a network QoS policy containing an ip-criteria statement to any object except a network IP interface will result in an error.

The no form of this command deletes all the entries specified under this node. Once IP criteria entries are removed from a network ingress policy, the IP criteria is removed from all network interfaces where that policy is applied. This command is supported on FP2- and higher-based hardware and is otherwise ignored.

IP criteria-based network ingress policies are used to select the appropriate ingress queue and corresponding forwarding class for matched traffic. This command is used to enter the context to create or edit policy entries that specify IPv6 criteria such as IP quintuple lookup or DiffServ code point.

The 7750 SR OS implementation will exit on the first match found and execute the actions in accordance with the accompanying action command. For this reason, entries must be sequenced correctly from most to least explicit.

The classification only applies to the outer IPv6 header of non-tunneled traffic.

Attempting to apply a network QoS policy containing an ipv6-criteria statement to any object except a network IP interface will result in an error.

The no form of this command deletes all the entries specified under this node. Once IP criteria entries are removed from a network ingress policy, the IP criteria is removed from all network interfaces where that policy is applied.

This command is supported on FP2- and higher-based hardware and is otherwise ignored.

This command is used to create or edit an IP or IPv6 criteria entry for the policy. Multiple entries can be created using unique entry-id numbers.

The list of flow criteria is evaluated in a top-down manner with the lowest entry ID at the top and the highest entry ID at the bottom. If the defined match criteria for an entry within the list matches the information in the ingress packet, the system stops matching the packet against the list and performs the matching entries reclassification actions. If none of the entries match the packet, the IP flow reclassification list has no effect on the packet.

An entry is not populated in the list unless the action command is executed for the entry. An entry that is not populated in the list has no effect on ingress packets. If the action command is executed without any explicit reclassification actions specified, the entry is populated in the list allowing packets matching the entry to exit the list, preventing them from matching entries lower in the list. Since this is the only flow reclassification entry that the packet matched and this entry explicitly states that no reclassification action is to be performed, the matching packet will not be reclassified.

The no form of this command removes the specified entry from the policy. Entries removed from the policy are immediately removed from all services where that policy is applied.

This mandatory command associates the forwarding class and packet profile with specific IP or IPv6 criteria entry ID.

Packets that meet all match criteria within the entry have their forwarding class and packet profile set based on the parameters included in the action parameters.

The action command must be executed for the match criteria to be added to the active list of entries.

Each time action is executed on a specific entry ID, the previously entered values for fc fc-name and profile are overridden with the newly defined parameters.

The no form of the command removes the entry from the active entry list. Removing an entry on a policy immediately removes the entry from all network interfaces using the policy. All previous parameters for the action are lost.

This command creates a context to configure match criteria for an ingress network QoS policy match criteria. When the match criteria have been satisfied, the action associated with the match criteria is executed.

If more than one match criteria (within one match statement) are configured, then all criteria must be satisfied (AND function) before the action associated with the match is executed.

A match context can consist of multiple match criteria, but multiple match statements cannot be entered per entry.

It is possible that a network QoS policy includes the dscp map command, the dot1p map command, and an IP match criteria. When multiple matches occur for the traffic, the order of precedence is used to arrive at the final action. The order of precedence is as follows:

802.1p bits

The no form of this command removes the match criteria for the entry-id.

This command creates a context to configure match criteria for a network QoS policy match IPv6 criteria. When the match criteria have been satisfied, the action associated with the match criteria is executed.

If more than one match criteria (within one match statement) are configured, all criteria must be satisfied (AND function) before the action associated with the match is executed.

A match context can consist of multiple match criteria, but multiple match statements cannot be entered per entry.

It is possible that a network ingress policy includes the dscp map command, the dot1p map command, and an IPv6 match criteria. When multiple matches occur for the traffic, the order of precedence is used to arrive at the final action. The order of precedence is as follows:

802.1p bits

The no form of this command removes the match criteria for the entry-id.

This command configures a DiffServ Code Point (DSCP) code point to be used as a network ingress QoS policy match criterion.

The no form of this command removes the DSCP match criterion.

This command configures a destination address range to be used as a network ingress QoS policy match criterion.

To match on the destination address, specify the address and its associated mask, for example, 10.1.0.0/16. The conventional notation of 10.1.0.0 255.255.0.0 can also be used.

The no form of this command removes the destination IP address match criterion.

This command configures a destination TCP or UDP port number or port range for a network ingress QoS policy match criterion.

The no form of this command removes the destination port match criterion.

This command configures fragmented or non-fragmented IP packets as a network ingress QoS policy match criterion.

The no form of this command removes the match criterion and matches all packets regardless of whether they are fragmented or not.

This command configures fragmented or non-fragmented IPv6 packets as a network ingress QoS policy match criterion.

The no form of this command removes the match criterion and matches all packets regardless of whether they are fragmented or not.

This command configures a source IPv4 or IPv6 address range to be used as a network ingress QoS policy match criterion.

To match on the source IPv4 or IPv6 address, specify the address and its associated mask, for example, 10.1.0.0/16. The conventional notation of 10.1.0.0 255.255.0.0 can also be used for IPv4.

The no form of the command removes the source IPv4 or IPv6 address match criterion.

This command configures a source TCP or UDP port number or port range for a network ingress QoS policy match criterion.

The no form of this command removes the source port match criterion.

This command renumbers existing QoS policy criteria entries to properly sequence policy entries.

This can be required in some cases since the router exits when the first match is found and executes the actions in accordance with the accompanying action command. This requires that entries be sequenced correctly from most to least explicit.

This command is used to enable tunnel QoS mapping on all ingress network IP interfaces that the network-qos-policy-id is associated with. The command may be defined at anytime after the network QoS policy has been created. Any network IP interfaces currently associated with the policy will immediately start to use the internal IP ToS field of any tunnel terminated IP routed packet received on the interface, ignoring any QoS markings in the tunnel portion of the packet.

This attribute provides the ability to ignore the network ingress QoS mapping of a terminated tunnel containing an IP packet that is to be routed to a base router or VPRN destination. This is advantageous when the mapping for the tunnel QoS marking does not accurately or completely reflect the required QoS handling for the IP routed packet. When the mechanism is enabled on an ingress network IP interface, the IP interface will ignore the tunnel’s QoS mapping and derive the internal forwarding class and profile based on the precedence or DiffServe Code Point (DSCP) values within the routed IP header ToS field compared to the Network QoS policy defined on the IP interface.

The default state is not to enforce tunnel termination IP routed QoS override within the network QoS policy.

The no form of the command removes tunnel termination IP routed QoS override from the network QoS policy and all ingress network IP interfaces associated with the policy.

This command creates a mapping between the LSP EXP bits of the network ingress traffic and the forwarding class.

Ingress traffic that matches the specified LSP EXP bits will be assigned to the corresponding forwarding class. Multiple commands can be entered to define the association of some or all eight LSP EXP bit values to the forwarding class. For undefined values, packets are assigned to the forwarding class specified under the default-action command.

The no form of this command removes the association of the LSP EXP bit value to the forwarding class. The default-action then applies to that LSP EXP bit pattern.

This command is used to enter the CLI node that creates or edits egress policy entries that specify the forwarding class queues to be instantiated when this policy is applied to the network port.

The forwarding class and profile state mapping to in- and out-of-profile DiffServ Code Points (DSCPs), dot1p, and MPLS EXP bits mapping for all labeled packets are also defined in this context.

All service packets are aggregated into DiffServ-based egress queues on the network interface. The service packets are transported either with IP GRE encapsulation or over a MPLS LSP. The exception is with the IES service. In this case, the actual customer IP header has the DSCP field mapped.

All out-of-profile service packets are marked with the corresponding out-of-profile DSCP, dot1p, or the EXP bit value at network egress. All the in-profile service ingress packets are marked with the corresponding in-profile DSCP, dot1p, or EXP bit value based on the forwarding class to which they belong. The exceed-profile traffic is marked with the same value as out-of-profile traffic and the inplus-profile traffic is marked with the same value as in-profile traffic.

This command is used to explicitly define the marking of the DE bit for fc fc-name according to the inplus-profile or in-profile and out-of-profile or exceed-profile status of the packet (fc-name may be used to identify the dot1p value).

This command is used whenever the dot1p bits are set to a common value regardless of the internal profile of the packets. Although it is not mandatory, this command should be used in combination with the de-mark command to enable the marking of the DE bit according to the internal profile of the packet.

This command acts as a shortcut for configuring the two existing commands with the same dot1p priority.

The dot1p dot1p-priority command is saved in the configuration as dot1p-in-profile dot1p-priority and dot1p-out-profile dot1p-priority. The inplus-profile traffic is marked with the same value as in-profile traffic. The exceed-profile traffic is marked with the same value as out-of-profile traffic.

This command specifies dot1p in-profile mappings. The inplus-profile traffic is marked with the same value as in-profile traffic.

The no form of the command resets the configuration to the default in-profile dot1p-priority setting for policy-id 1.

This command specifies dot1p out-of-profile mappings.

The exceed-profile traffic is marked with the same value as out-of-profile traffic.

The no form of the command resets the configuration to the default out-profile dot1p-priority setting for policy-id 1.

This command specifies the in-profile DSCP name for the forwarding class. The corresponding DSCP value will be used for all IP packets that require marking at egress on this forwarding class queue, and that are in-profile. The inplus-profile traffic is marked with the same value as in-profile traffic.

When multiple DSCP names are associated with the forwarding class at network egress, the last name entered will overwrite the previous value.

The no form of this command resets the configuration to the factory default in-profile DSCP name setting for policy-id 1.

This command specifies the out-of-profile DSCP name for the forwarding class. The corresponding DSCP value will be used for all IP packets requiring marking the egress on this forwarding class queue that are out-of-profile.The exceed-profile traffic is marked with the same value as out-of-profile traffic.

When multiple DSCP names are associated with the forwarding class at network egress, the last name entered will overwrite the previous value.

The no form of this command resets the configuration to the factory default out-of-profile DSCP name setting for policy-id 1.

This command specifies the in-profile LSP EXP value for the forwarding class. The EXP value will be used for all LSP labeled packets requiring marking that require marking at egress on this forwarding class queue, and that are in-profile. The inplus-profile traffic is marked with the same value as in-profile traffic.

When multiple EXP values are associated with the forwarding class at network egress, the last name entered will overwrite the previous value.

The no form of this command resets the configuration to the factory default in-profile EXP setting.

This command specifies the out-of-profile LSP EXP value for the forwarding class. The EXP value will be used for all LSP labeled packets that require marking at egress on this forwarding class queue, and that are out-of-profile. The exceed-profile traffic is marked with the same value as out-of-profile traffic.

When multiple EXP values are associated with the forwarding class at network egress, the last name entered will overwrite the previous value.

The no form of this command resets the configuration to the factory default out-of-profile EXP setting.

This command is used to redirect the FC of a packet of a pseudowire (PW) or network IP interface to an egress port queue group.

It defines the mapping of an FC to a queue ID or a policer ID and a queue ID, and redirects the lookup of the queue or policer of the same ID in some egress port queue-group instance. However, the queue-group name and instance are explicitly provided only at the time the network QoS policy is applied to egress context of a spoke-sdp or a network IP interface.

The no version of this command removes the redirection of the FC.

This command configures an IP Differentiated Services Code Point (DSCP) value that must be matched to perform the associated reclassification actions. If an egress packet on an IES/VPRN interface spoke SDP, on a CSC network interface in a VPRN, or on a network interface that the network QoS policy is applied to, matches the specified IP DSCP value, the forwarding class and profile may be overridden.

By default, the forwarding class and profile of the packet are derived from ingress classification and profiling functions. Matching a DHCP-based reclassification rule will override all IP precedence-based reclassification rule actions.

The IP DSCP bits used to match against DSCP reclassification rules come from the Type of Service (ToS) field within the IPv4 header or the Traffic Class field from the IPv6 header. If the packet does not have an IP header, DSCP-based matching is not performed.

The IP precedence- and DSCP-based reclassification are supported on a network interface, on a CSC network interface in a VPRN, and on a PW used in an IES or VPRN spoke interface. The CLI will block the application of a network QoS policy with the egress reclassification commands to the spoke SDP part of a Layer 2 service.

Conversely, the CLI will not allow the user to add the egress reclassification commands to a network QoS policy if the policy is being used by a Layer 2 spoke SDP.

The egress reclassification commands will only take effect if the redirection of the spoke SDP or CSC interface to use an egress port queue group succeeds. For example, the following CLI command would be successful:

config>service>vprn>if>spoke-sdp>egress>qos network-policy-id port-redirect-group queue-group-name instance instance-id

config>service>ies>if>spoke-sdp>egress>qos network-policy-id port-redirect-group queue-group-name instance instance-id

config>service>vprn>nw-if>qos network-policy-id port-redirect-group queue-group-name instance instance-id

If the redirection command fails, the PW will use the network QoS policy assigned to the network IP interface, however any reclassification in the network QoS policy applied to the network interface will be ignored.

The no version of this command removes the egress reclassification rule.

This command defines a specific IP precedence value that must be matched in order to perform the associated reclassification actions. If an egress packet on an IES/VPRN interface spoke SDP, on a CSC network interface in a VPRN, or network interface that the network QoS policy is applied to, matches the specified IP precedence value, the forwarding class and profile may be overridden.

By default, the forwarding class and profile of the packet is derived from ingress classification and profiling functions.

The IP precedence bits used to match against the reclassification rules come from the Type of Service (ToS) field within the IPv4 header or the Traffic Class field from the IPv6 header. If the packet does not have an IP header, IP precedence-based matching is not performed.

The IP precedence- and DSCP-based reclassification are supported on a network interface, on a CSC network interface in a VPRN, and on a PW used in an IES or VPRN spoke interface.

This command will block the application of a network QoS policy with the egress reclassification commands to a spoke SDP part of a Layer 2 service. Conversely, this command will not allow the user to add the egress reclassification commands to a network QoS policy if it is being used by a Layer 2 spoke SDP.

The egress reclassification commands will only take effect if the redirection of the spoke SDP or CSC interface to use an egress port queue-group succeeds. For example, the following commands will succeed:

When the redirection command fails in CLI, the PW will use the network QoS policy assigned to the network IP interface; however, any reclassification in the network QoS policy applied to the network interface will be ignored.

The no version of this command removes the egress reclassification rule.

This command remarks both customer traffic and egress network IP interface traffic; VPRN customer traffic is not remarked. The remarking is based on the forwarding class to DSCP and LSP EXP bit mapping defined under the egress node of the network QoS policy.

Normally, packets that ingress on network ports have either the DSCP or, for MPLS packets, LSP EXP bit set by an upstream router. The packets are placed in the appropriate forwarding class based on the DSCP-to-forwarding class mapping or the LSP EXP-to-forwarding class mapping. The DSCP or LSP EXP bits of such packets are not altered as the packets egress this router, unless remarking is enabled.

Remarking can be required if this router is connected to a different DiffServ domain where the DSCP-to-forwarding class mapping is different.

Normally, no remarking is necessary when all router devices are in the same DiffServ domain.

The network QoS policy supports an egress flag that forces remarking of packets that were received on trusted IES and network IP interfaces. This provides the capability of remarking without regard to the ingress state of the IP interface on which a packet was received. The effect of the setting of the egress network remark trusted state on each type of ingress IP interface and trust state is shown in the following table.

The remark trusted state has no effect on packets received on an ingress VPRN IP interface.

The no form of this command resets the configuration to the default behavior.

This command enters the context to configure DSCP/dot1p remarking for self-generated traffic.

This command configures DSCP/dot1p remarking for self-generated application traffic. When an application is configured using this command, the specified DSCP name/value is used for all packets generated by this application within the router instance it is configured. The instances can be base router, vprn, or management.

Using the value configured in this command:

Only one DSCP name/value can be configured per application, if multiple entries are configured, the subsequent entry overrides the previous configured entry.

The no form of this command reverts back to the default value.

This command creates a mapping between the DiffServ Code Point (DSCP) of the self-generated traffic and the forwarding class.

Self-generated traffic that matches the specified DSCP will be assigned to the corresponding forwarding class. Multiple commands can be entered to define the association of some or all 64 DSCPs to the forwarding class.

All DSCP names that define a DSCP value must be explicitly defined

The no form of this command removes the DSCP-to-forwarding class association.