Abstract: A method for controlling a session for providing a certain service, at least one data stream being associated with the session. The method includes: obtaining a session control policy from a session control message, or generating such a policy as a function of a type of the service, which session control policy is to be implemented in the event of at least one data stream being inactive, i.e. in the event of a data stream associated with the session not being set up, or being released prematurely; sending the session control policy in a session control message to a control entity in charge of implementing the session control policy; and on detecting inactivity of at least one data stream, implementing the session control policy.

Abstract: The invention relates to a method for splitting application messages in an IP (Internet Protocol) network, comprising the following steps: a) a node of the IP network, termed the first node, receives a message having to be transmitted to another node of the IP network, termed the second node; b) said first node compares the length of said message with the maximum transmission size MTU (Maximum Transmission Unit) characterizing the path between said first node and said second node; and if said length is greater than the MTU, c) the first node splits the message into at least two segments, and inserts each respective segment into a respective data packet of length less than or equal to the MTU; d) the first node transmits said data packets to the second node; and e) the second node assembles said segments so as to reconstruct said message.

Abstract: A method of dynamically updating DNS information in an IP network. A first entity of the IP network, distinct from DNS servers of the IP network, receives a request from a second entity of the network, likewise distinct from the DNS servers but possessing a copy of DNS information, constituting a record in at least one DNS server of the IP network. The first entity responds to the request by sending a message to the second entity, including an indication about the need to modify the copy. On receiving the message, the second entity determines whether the indication includes at least one additional instruction. If so, the second entity: performs a DNS interrogation, and on receiving a response, modifies its copy of the DNS information based the response and the additional instruction; or modifies its copy of the DNS information based on the additional instruction without performing a DNS interrogation.

Abstract: A method for controlling a session for providing a certain service, at least one data stream being associated with the session. The method includes: obtaining a session control policy from a session control message, or generating such a policy as a function of a type of the service, which session control policy is to be implemented in the event of at least one data stream being inactive, i.e. in the event of a data stream associated with the session not being set up, or being released prematurely; sending the session control policy in a session control message to a control entity in charge of implementing the session control policy; and on detecting inactivity of at least one data stream, implementing the session control policy.

Abstract: A method of dynamically updating DNS information in an IP network. A first entity of the IP network, distinct from DNS servers of the IP network, receives a request from a second entity of the network, likewise distinct from the DNS servers but possessing a copy of DNS information, constituting a record in at least one DNS server of the IP network. The first entity responds to the request by sending a message to the second entity, including an indication about the need to modify the copy. On receiving the message, the second entity determines whether the indication includes at least one additional instruction. If so, the second entity: performs a DNS interrogation, and on receiving a response, modifies its copy of the DNS information based the response and the additional instruction; or modifies its copy of the DNS information based on the additional instruction without performing a DNS interrogation.

Abstract: A processing method for use in a communications network to process a service request relating to a user. During a first stage referred to as a “detection” stage an interrogation server interrogates a subscriber server in order to obtain an identifier of a nominal registration server to which it transfers the service request, and in the event of the transfer failing, it interrogates the subscriber server in order to obtain the capabilities that a registration server needs to possess in order to be capable of delivering the services authorized for the user. Once the nominal registration server is considered to be unreachable, a second stage is started, referred to as an “optimization” stage, during which a single transfer is made to registration servers of the service request relating to a user.

Abstract: A method and apparatus are provided for selecting a network resource, in which a “controlled-entity” device connected to an IP network (20) is also capable of accessing at least one other physical and/or virtual telecommunications network. According to the method, after a step in which a request is sent or relayed by the controlled-entity or is sent to the controlled-entity, the controlled-entity then receives a session control signal containing a resource identifier representative of the physical or virtual network that the controlled-entity may or must use in order to satisfy the request.