Cisco Cisco ASR 5000 Technical Manual
Policy Overload Redirect
This should be configured if a redirect action is configured for the Call Session Control Function (CSCF),
HRPD Serving Gateway (HSGW), Home Agent (HA), or PDSN service.
HRPD Serving Gateway (HSGW), Home Agent (HA), or PDSN service.
The CSCF service has this command configured under the CSCF policy rules configuration.
•
The HSGW service, HA service, and PDSN service has this command configured under the
respective service configurations.
respective service configurations.
•
Congestion Control Policy for MME Service
Prior to Release 14.0, the congestion control policy for the MME service can be defined similarly to the CLI
syntax that is mentioned in the previous section, but with some additional options. Here is the CLI syntax:
syntax that is mentioned in the previous section, but with some additional options. Here is the CLI syntax:
congestion−control policy mme−service action { drop | none | reject | report−overload
{ permit−emergency−sessions | reject−new−sessions | reject−non−emergency−sessions }
enodeb−percentage <percentage> }
In addition to the drop, none, and reject actions, the MME service also has the option to report overload
conditions for the eNodeBs. The MME invokes the S1 overload procedure with the S1AP Overload Start
message in order to report an overload condition to the specified proportion of eNodeBs to which the MME
has an S1 interface connection. The MME selects the eNodeBs at random. Two overloaded MMEs in the
same pool do not send overload messages to the same eNodeBs. When the MME has recovered and can
increase its load, then it sends an S1AP Overload Stop message. In addition, these actions can be completed
when a report overload action is configured:
conditions for the eNodeBs. The MME invokes the S1 overload procedure with the S1AP Overload Start
message in order to report an overload condition to the specified proportion of eNodeBs to which the MME
has an S1 interface connection. The MME selects the eNodeBs at random. Two overloaded MMEs in the
same pool do not send overload messages to the same eNodeBs. When the MME has recovered and can
increase its load, then it sends an S1AP Overload Stop message. In addition, these actions can be completed
when a report overload action is configured:
Permit−emergency−sessions: This action allows only emergency sessions on the MME during an
overload period.
overload period.
•
Reject−new−sessions: This action causes a rejection of all new sessions inbound towards the MME
during an overload situation.
during an overload situation.
•
Reject−non−emergency−sessions: This action causes all non−emergency sessions to be rejected on
the MME during an overload period.
the MME during an overload period.
•
Enodeb−percentage: This action configures the percentage of known eNodeBs that receive the
overload report. The percentage can range between one and one hundred.
overload report. The percentage can range between one and one hundred.
•
In Releases 14.0 and later, the MME service can have three different policies and associated action profiles.
Here is the CLI syntax:
Here is the CLI syntax:
congestion−control policy { critical mme−service action−profile <action_profile_name> |
major mme−service action−profile <action_profile_name> | minor mme−service
action−profile <action_profile_name> }
There are three policy types that can be configured for the MME in Releases 14.0 and later:
Critical: This defines the critical congestion control threshold for the MME service.
•
Major: This defines the major congestion control threshold for the MME service.
•
Minor: This defines the minor congestion control threshold for the MME service.
•