[Contents]
[Prev]
[Next]
[Index]
[Report an Error]
Controlling ServiceChange Commands Sent from the VPG to the
PGC
For seamless interoperability between the JUNOS packet gateway
and PGC devices, you can control the method and reason that the VPG
includes in ServiceChange commands that it sends to the PGC. You can
also prevent the system from being overloaded with messages for certain
state changes by specifying that the VPG not send a request or notification
when those changes occur.
You can specify the method and reason that the VPG includes
in ServiceChange commands when the state of one of the following changes:
- Control associations
- Virtual interfaces
- Contexts
Control Association States
A control association is a relationship where the PGC is controlling
the VPG. Each VPG has only one control association at any time.
Table 10 describes
the control association states.
Table 10: Control Association
States
| Control Association State | Description |
|
Disconnected
|
The control association is in the Disconnected state. No PGC
is controlling the VPG, and incoming H.248 messages are ignored. The
control association remains disconnected as long as the VPG is Out-of-Service.
Depending on what caused the VPG to become Out-of-Service, the
VPG either drops H.248 commands or answers them with a port-unreachable
ICMP error.
|
|
Connecting
|
The control association is in Connecting state between the time
the VPG sends a registration request to the PGC and the time the PGC
accepts, rejects, or aborts the request.
The VPG rejects incoming H.248 commands while the control association
is in the Connecting state with error # 505: “Transaction
Request Received before a ServiceChange Reply has been received”.
|
|
Draining
|
The control association enters the Draining state when an administrator
instructs the VPG to gracefully transition from In-Service to Out-of-Service.
The PGC transitions to Out-of-Service when the controlling PGC subtracts
all of the VPG’s H.248 terminations.
The VPG accepts only Subtract and AuditValue commands from the
controlling PGC. It rejects all other commands with error # 502:
“Not Ready”.
|
When the state of a control association changes, the VPG can
send the following types of ServiceChange commands to the PGC:
- Registration Requests—The VPG sends a Registration
Request ServiceChange command to request that a PGC become its controlling
PGC. The VPG sends these requests when a control association enters
the Connecting state.
- Unregistration Messages—The VPG sends an Unregistration
ServiceChange command to its controlling PGC when it transitions to
the Out-of-Service (Disconnected) service state because of an administration
operation or a failure. The failure can be the result of a MultiServices
PIC or Flexible PIC Concentrator (FPC) failure or because the MultiServices
PIC was powered off or removed.
- Notification Messages—The VPG sends a notification
ServiceChange command to its controlling PGC when the control association
transitions between the Connected and Draining states and vice versa.
Method and Reason Options for Control Association
State Changes
You can control the ServiceStateMethod and ServiceStateReason
that the VPG includes in ServiceChange commands for control associations.
You can use the CLI to specify the method and reason that the
VPG includes in ServiceChange commands for control associations. Table 11 shows
the method and reason options available for each reported state and
the events that led to the report.
Table 11: Options
for Method and Reason in ServiceChange Commands for Control Associations
| Reported Association State | Event Leading to Report | Options | Embedded H.248 Reason | Explanation |
|
Disconnect
|
Controller failure
|
FL/909
|
PGC impending failure
|
VPG is reregistering with a new PGC following a disconnection
of the VPG and PGC.
|
|
RS/902
|
Warm boot
|
VPG is reregistering with a new PGC following a disconnection
of the VPG and PGC.
|
|
Reconnect
|
DC/900
|
Service restored
|
VPG is registering with the last controlling PGC following a
disconnection of the VPG and PGC.
|
|
RS/902
|
Warm boot
|
VPG is transitioning to In-Service, and the previously installed
state is retained.
|
|
Down
|
Administrative
|
FO/905
|
Termination taken out of service
|
VPG is transitioning to Out-of-Service because of an administrative
operation.
|
|
FO/908
|
VPG impending failure
|
VPG root termination transitioned to Out-of-Service and is unable
to process request.
|
|
none
| |
No message is sent for this event.
|
|
Failure
|
FO/904
|
Termination malfunctioning
|
VPG is transitioning to Out-of-Service because of a failure.
|
|
FO/908
|
VPG impending failure
|
VPG root termination transitioned to Out-of-Service because
of a failure.
|
|
none
| |
No message is sent for this event.
|
|
Graceful
|
GR/905
|
Termination taken out of service
|
The control association entered the Draining state because of
an administrative operation.
|
|
none
| |
No message is sent for this event.
|
|
Up
|
Cancel graceful
|
RS/908
|
Cancel graceful
|
The control association transitioned from the Draining state
to the Forwarding state.
|
|
none
| |
No message is sent for this event.
|
|
Cold failover
|
FL/920
|
Cold failover
|
VPG is registering following a graceful Routing Engine switchover.
The previously installed state is reset.
|
|
RS/901
|
Cold boot
|
VPG is transitioning to In-Service. The previously installed
state is not retained.
|
|
Warm failover
|
FL/919
|
PGC impending failure
|
VPG is registering with a new PGC following a disconnection
of the VPG and PGC.
|
|
RS/902
|
Warm boot
|
VPG is transitioning to In-Service, and the previously installed
state is retained.
|
Configuring the Method and Reason in ServiceChange
Commands for Control Associations
Step-by-Step Procedure
To configure the method and reason in ServiceChange commands
for control associations:
- Access the configuration of the service change
control association indications properties.
- [edit services pgcp]
- user@host#edit gateway vpg-1 h248-options
service-change control-association-indications
- Specify the method and reason that the
VPG includes in Registration Request ServiceChange commands when it
attempts to reregister with the PGC or register with a new PGC after
the control association is disconnected.
- [edit services pgcp gateway vpg-1 h248-options service-change
control-association-indications]
- user@host#set disconnect controller-failure
restart-902
- Specify the method and reason that the
VPG includes in Registration Request ServiceChange commands when it
attempts to reregister with the PGC or register with a new PGC after
the control association is disconnected.
- [edit services pgcp gateway vpg-1 h248-options service-change
control-association-indications]
- user@host#set disconnect reconnect restart-902
- Specify the method and reason that the
VPG includes in Unregistration Messages in ServiceChange commands
that it sends to the PGC when a control association transitions to
Out-of-Service because of an administrative operation.
- [edit services pgcp gateway vpg-1 h248-options service-change
control-association-indications]
- user@host#set down administrative forced-908
- Specify the method and reason that the
VPG includes in Unregistration Messages in ServiceChange commands
that it sends to the PGC when a control association transitions to
Out-of-Service because of a failure.
- [edit services pgcp gateway vpg-1 h248-options service-change
control-association-indications]
- user@host#set down failure forced-904
- Specify the method and reason that the
VPG includes in Notification ServiceChange commands that it sends
to the PGC when the control association transitions from In-Service
to Out-of-Service-Graceful.
- [edit services pgcp gateway vpg-1 h248-options service-change
control-association-indications]
- user@host#set down graceful graceful-905
- Specify the method and reason that the
VPG includes in Notification ServiceChange commands that it sends
to the PGC when the control association has returned to the Connected
state.
- [edit services pgcp gateway vpg-1 h248-options service-change
control-association-indications]
- user@host#set up cancel-graceful restart-918
- Specify the method and reason that the
VPG includes in Registration ServiceChange commands when it attempts
to register with a new PGC following a cold failover.
- [edit services pgcp gateway vpg-1 h248-options service-change
control-association-indications]
- user@host#set up failover-cold failover-920
- Specify the method and reason that the
VPG includes in Registration ServiceChange commands when it attempts
to register with a new PGC following a warm failover.
- [edit services pgcp gateway vpg-1 h248-options service-change
control-association-indications]
- user@host#set up failover-warm restart-902
Virtual Interface States
Table 12 describes the
virtual interface states.
Table 12: Virtual Interface States
| Virtual Interface Operational
State | Description |
|
Blocked
|
A virtual interface is in the Blocked state when the interface
is Out-of-Service.
While a virtual interface is in the Blocked state, all VPGs
do not add new terminations using the interface. Likewise, the VPGs
rejects H.248 commands other than Subtract and AuditValue commands
on existing terminations on the interface. The error that the VPG
returns for commands that it rejects depends on the reason that caused
the virtual interface to be Out-of-Service:
-
Error #502: "Not ready"—If the virtual
interface is Out-of-Service because of an administrative operation.
-
Error #529: "Internal hardware failure in PG"—If the virtual interface is Out-of-Service because of a failure.
|
|
Forwarding
|
A virtual interface is in the Forwarding state when it is functioning
normally. All gates are using the interface process data flows according
to the H.248 properties installed on them.
|
|
Draining
|
A virtual interface enters the Draining state when an administrator
instructs the virtual interface to gracefully transition from In-Service
to Out-of-Service. The virtual interface automatically transitions
to Out-of-Service when it is no longer used by any termination in
any of the VPGs.
A virtual interface that is in the Draining state is In-Service
and existing gates process data flows normally. However, as in the
Blocked state, the VPGs do not add new terminations using that virtual
interface or to perform any command other than Subtract and AuditValue
on existing terminations on the interface. If the VPG receives other
commands, it replies with error #502: "Not ready".
|
When the state of a virtual interface changes, the VPG can send
the following types of ServiceChange commands to the PGC:
- Service-Restoration—The VPG sends Service-Restoration
ServiceChange commands when a virtual interface transitions to the
In-Service service state; that is, it transitions from the Blocked
state to the Forwarding operational state.
- Service-Interruption—The VPG sends Service-Interruption
ServiceChange commands when a virtual interface transitions to the
Out-of-Service, or Blocked, service state.
- Notification Messages—The VPG sends Notification
ServiceChange commands when a virtual interface transitions between
the Forwarding and Draining states and vice versa.
Method and Reason Options for Virtual Interface
State Changes
You can control the ServiceStateMethod and ServiceStateReason
that the VPG includes in ServiceChange commands for virtual interface
state changes.
You can use the CLI to specify the method and reason that the
VPG includes in ServiceChange commands for virtual interfaces. Table 13 explains
the method and reason options available for each reported state and
the events that led to the report.
Table 13: Options
for Method and Reason in ServiceChange Commands for Virtual Interfaces
| Reported State | Event Leading to Report | Options | Embedded H.248 Reason | Explanation |
|
Virtual interface down
|
Administrative
|
FO/905
|
Termination taken out of service
|
Virtual interface is transitioning to Out-of-Service because
of an administrative operation.
|
|
FO/906
|
Loss of lower-layer connectivity
|
Virtual interface is transitioning to Out-of-Service because
of a loss of layer 2 connectivity caused by the logical or physical
interface being administratively disabled.
|
|
none
| |
No message is sent for this event.
|
|
Failure
|
FO/904
|
Termination malfunctioning
|
Virtual interface is transitioning to Out-of-Service because
of an internal failure.
|
|
FO/906
|
Loss of lower-layer connectivity
|
Virtual interface is transitioning to Out-of-Service because
of a loss of layer 2 connectivity on the logical or physical interface.
|
|
none
| |
No message is sent for this event.
|
|
Graceful
|
GR/905
|
Termination taken out of service
|
Virtual interface has entered the Draining state.
|
|
none
| |
No message is sent for this event.
|
|
Link loss
|
FO/906
|
Loss of lower-layer connectivity
|
Virtual interface is transitioning to Out-of-Service because
of a loss of layer 2 connectivity on the logical or physical interface.
|
|
none
| |
No message is sent for this event.
|
|
Virtual interface up
|
Cancel graceful
|
RS/918
|
Cancel graceful
|
Virtual interface has returned to the Forwarding state.
|
|
none
| |
No message is sent for this event.
|
|
warm
|
RS/900
|
Service restored
|
Virtual interface has become In-Service and is in the Forwarding
state.
|
|
none
| |
No message is sent for this event.
|
Configuring the Method and Reason in ServiceChange
Commands for Virtual Interfaces
Step-by-Step Procedure
To configure the method and reason in ServiceChange commands
for virtual interfaces:
- Access the configuration of the service change
virtual interface indications properties.
- [edit services pgcp]
- user@host#edit gateway vpg-1 h248-options
service-change virtual-interface-indications
- Specify the method and reason that the
VPG includes in Service-Interruption ServiceChange commands that it
sends to the PGC when a virtual interface changes to Out-of-Service
because of an administrative operation.
- [edit services pgcp gateway vpg-1 h248-options service-change
virtual-interface-indications]
- user@host#set virtual-interface-down administrative
forced-906
- Specify the method and reason that the
VPG includes in Service-Interruption ServiceChange commands that it
sends to the PGC when a virtual interface transitions to Out-of-Service
because of a failure.
- [edit services pgcp gateway vpg-1 h248-options service-change
virtual-interface-indications]
- user@host#set virtual-interface-down failure
forced-906
- Specify the method and reason that the
VPG includes in Notification ServiceChange commands that it sends
to the PGC when a virtual interface transitions between the Forwarding
and Draining states.
- [edit services pgcp gateway vpg-1 h248-options service-change
virtual-interface-indications]
- user@host#set virtual-interface-down graceful
none
- Specify the method and reason that the
VPG includes in Service-Interruption ServiceChange commands that it
sends to the PGC when the virtual interface transitions to Out-of-Service
because of a link loss.
- [edit services pgcp gateway vpg-1 h248-options service-change
virtual-interface-indications]
- user@host#set virtual-interface-down link-loss
forced-906
- Specify the method and reason that the
VPG includes in Notification ServiceChange commands that it sends
to the PGC when the virtual interface transitions from the Draining
state to the Forwarding state.
- [edit services pgcp gateway vpg-1 h248-options service-change
virtual-interface-indications]
- user@host#set virtual-interface-up cancel-graceful
restart-918
- Specify the method and reason that the
VPG includes in Service-Restoration ServiceChange commands that it
sends to the PGC when a virtual interface transitions to In-Service.
- [edit services pgcp gateway vpg-1 h248-options service-change
virtual-interface-indications]
- user@host#set virtual-interface-up warm restart-900
Context States
The VPG sends context Service-Interruption messages when the
gates of a specific context no longer provide their configured service.
When such a message is issued, both terminations included in the context
become Out-of-Service.
You can use the CLI to specify the method and reason that the
VPG includes in Service-Interruption ServiceChange commands that it
sends to the PGC when a state loss occurs.Table 14 describes the method
and reason options available.
Table 14: Options
for Method and Reason in ServiceChange Commands for Specific Contexts
| Reported State | Event Leading to Report | Options | Embedded H.248 Reason | Explanation |
|
State loss
|
Mismatch between pgcpd process and MultiServices PIC states
|
FO/910
|
State loss because of a media failure
|
A mismatch between the pgcpd process and the MultiServices PIC
states on one or more of the context’s gates.
|
|
FO/915
|
State loss
|
A mismatch between the pgcpd process and the MultiServices PIC
states was detected on one or more of the context’s gates.
|
|
none
| |
No message is sent for this event.
|
Configuring the Method and Reason in ServiceChange
Commands for Contexts
Step-by-Step Procedure
To configure the method and reason in ServiceChange commands
for contexts:
- Access the configuration of the service change
context indications properties.
- [edit services pgcp]
- user@host#edit gateway vpg-1 h248-options
service-change context-indications
- Specify the method and reason that the
VPG includes in Service-Interruption ServiceChange commands that it
sends to the PGC after a state loss on a specific context.
- [edit services pgcp gateway vpg-1 h248-options service-change
context-indications]
- user@host#set state-loss forced-915
[Contents]
[Prev]
[Next]
[Index]
[Report an Error]
