Tech-invite   World Map
3GPPspecs     Glossaries     IETF     RFCs     Groups     SIP     ABNFs
Top        in Index        Prev        Next

TR 23.820 (CT4)
Study of
IMS restoration procedures

ToC      3GPP‑Page      Help       
V9.0.0 (Wzip)    2009/09    43 p.
V8.0.1    2008/12    42 p.

Rapporteur:  Mr. Blanco, German
See also:  IMS-Restoration-related TS/TR    

Although network nodes in the IMS Core Network should have a very high availability, some maintenance downtime and occasional failures are unavoidable. Communication links although designed with robust protocols between the network elements are also subject to failures. A set of standardized procedures for automatic restoration after loss or corruption of data could reduce the impact of these problems resulting in the improved service to the users. The intention is that similar cases as in TS 23.007 for the CS and PS Domains are covered also for the IMS domain.

This TR identifies the changes required in the 3GPP IMS specifications so that a consistent state is restored in the IMS Core Network, after, or during a planned, or unplanned stop of a network element. The study will go through the following steps:
  • Establish the requirements that should be covered with these procedures. That is which are the impacts to the end user service that are acceptable and which are not, after a network failure.
  • List the service interruption scenarios that need to be studied.
  • Provide solutions, so that in all the service interruption scenarios listed, the impacts to the end user service comply with the requirements. These solutions provide procedures for the automatic restoration to a consistent state in the network and indicate how to trigger these procedures.
  • Analyze the impacts of the solutions in the current specifications.
  • Conclusion and recommended way forward.
It is important to realise that these procedures are meant to be operational procedures for restoration and so care must be taken with what is existing and will exist with OA&M procedures to avoid overlap which could cause clashes.


Here        Top


1 Scope    2 References    3 Definitions, symbols and abbreviations    3.1 Definitions    3.2 Abbreviations    4 Requirements analysis and assumptions    4.1 Introduction    4.2 Persistency Requirements for Data    4.3 Impacts on Established Sessions    4.4 Impacts on Session Establishment Time    4.5 Required Manual Intervention    4.6 Loss of service    4.7 Avoidance of massive signalling    4.8 Load balancing    5 Service Interruption scenarios    5.1 Introduction    5.2 S-CSCF Service Interruption    5.3 P-CSCF Service Interruption    5.4 IMS-UE Service Interruption    5.5 SIP-AS Service Interruption    5.6 IP-CAN Service Interruption    5.7 HSS Service Interruption    6 Alternative solutions    6.1 Backup of S-CSCF Information in the HSS    6.2 Triggering of initial registration from the S-CSCF or P-CSCF    6.2A Precautionary de-registration of un-registered users    6.2B S-CSCF re-assignment for unregistered user    6.3 Second P-CSCF and deregistration from S-CSCF    6.4 Monitoring P-CSCF Health    6.5 Possible Solution for SIP-AS Service Restoration    6.6 Update of S-CSCF Name in the HSS after Loss of Data    6.7 Forking Service Restoration    6.8 Possible Solutions for SIP-AS Service Restoration    6.9 AS Behaviour After HSS Recovery    6.10 HSS Failover with no loss of service    7 Conclusions and recommendations    7.1 S-CSCF Service Interruption    7.2 S-CSCF Re-Selection during Re-Registration    7.3 SIP-AS Service Interruption    7.4 HSS Service Interruption    7.5 P-CSCF Service Interruption    A Change history   


1   Scope   Word-p. 6
2   References
3   Definitions, symbols and abbreviations
4   Requirements analysis and assumptions
5   Service Interruption scenarios
6   Alternative solutions
6.1   Backup of S-CSCF Information in the HSS      Up
6.2   Triggering of initial registration from the S-CSCF or P-CSCF   Word-p. 25
6.2A   Precautionary de-registration of un-registered users   Word-p. 28
6.2B   S-CSCF re-assignment for unregistered user
6.3   Second P-CSCF and deregistration from S-CSCF
6.4   Monitoring P-CSCF Health   Word-p. 31
6.5   Possible Solution for SIP-AS Service Restoration   Word-p. 34
6.6   Update of S-CSCF Name in the HSS after Loss of Data   Word-p. 35
6.7   Forking Service Restoration   Word-p. 37
6.8   Possible Solutions for SIP-AS Service Restoration   Word-p. 38
6.9   AS Behaviour After HSS Recovery   Word-p. 39
6.10   HSS Failover with no loss of service   Word-p. 40      Up
7   Conclusions and recommendations
A   Change history   Word-p. 43

Up        Top