Abstract: A computer implemented method, apparatus, and computer usable program code for processing data packets. A set of data fragments are received at the data processing system to form a set of received data fragments. Assembly of the set of data fragments is initiated into a data packet, and a determination as to whether the data packet is an incomplete data packet. Responsive to a determination that the data packet is an incomplete data packet, the incomplete data packet is filled with at least one character to form a final data packet. The final data packet is forwarded to the target.

Abstract: A computer implemented method, apparatus, and computer usable program code for processing data packets. A set of data fragments are received at the data processing system to form a set of received data fragments. Assembly of the set of data fragments is initiated into a data packet, and a determination as to whether the data packet is an incomplete data packet. Responsive to a determination that the data packet is an incomplete data packet, the incomplete data packet is filled with at least one character to form a final data packet. The final data packet is forwarded to the target.

Abstract: A computer implemented method, apparatus, and computer usable program code for processing data packets. A set of data fragments are received at the data processing system to form a set of received data fragments. Assembly of the set of data fragments is initiated into a data packet, and a determination as to whether the data packet is an incomplete data packet. Responsive to a determination that the data packet is an incomplete data packet, the incomplete data packet is filled with at least one character to form a final data packet. The final data packet is forwarded to the target.

Abstract: Methods, system, and products are disclosed for administering requests for data communications connections in a wide area network that include receiving, according to a data communications protocol, by a receiver from a sender, a request for a data communications connection, wherein the sender has a network address on a sender's network and the receiver has a network address on a receiver's network; characterizing the sender as local with respect to the receiver's network if the sender's network is the receiver's network; characterizing the sender as remote with respect to the receiver's network if the sender's network is not the receiver's network; and responding to the request in dependence upon a value of a location response flag and in dependence upon the characterization of the sender's network with respect to the receiver's network.

Abstract: Responsive to receiving a request for a data communications connection, a data communications module of a receiver determines whether a sender of the request is local or remote with respect to the receiver. The data communications module also determines whether a location response flag indicates local or remote. If the relative location of the sender with respect to the receiver corresponds with the location response flag, then the data communications connection is allowed. Otherwise, the connection is not allowed.

Abstract: The reassembly timer in a TCP/IP receiver is dynamically set to avoid data corruption on fast networks caused by IP identification wrap-around. A sender generates IP packets with a special option specifying a reassembly time for fragments based on the subnet of the destination or on the calculated round-trip time of the connection. A receiver sets the reassembly timer to the reassembly time specified in the option. A sender can generate IP packets with an alternative option setting a time stamp for the generated packet. A receiver calculates a traversal time based on the difference between the arrival time of the fragment and this time stamp, and sets the reassembly timer based on this traversal time. A receiver can independently set the reassembly timer to a first reassembly time set by the user for fast networks or to a second reassembly timer set by the user for slow networks based on whether the receiver and sender are on the same subnet.

Abstract: The reassembly timer in a TCP/IP receiver is dynamically set to avoid data corruption on fast networks caused by IP identification wrap-around. A sender generates IP packets with a special option specifying a reassembly time for fragments based on the subnet of the destination or on the calculated round-trip time of the connection. A receiver sets the reassembly timer to the reassembly time specified in the option. A sender can generate IP packets with an alternative option setting a time stamp for the generated packet. A receiver calculates a traversal time based on the difference between the arrival time of the fragment and this time stamp, and sets the reassembly timer based on this traversal time.

Abstract: The reassembly timer in a TCP/IP receiver is dynamically set to avoid data corruption on fast networks caused by IP identification wrap-around. A sender generates IP packets with a special option specifying a reassembly time for fragments based on the subnet of the destination or on the calculated round-trip time of the connection. A receiver sets the reassembly timer to the reassembly time specified in the option. A sender can generate IP packets with an alternative option setting a time stamp for the generated packet. A receiver calculates a traversal time based on the difference between the arrival time of the fragment and this time stamp, and sets the reassembly timer based on this traversal time. A receiver can independently set the reassembly timer to a first reassembly time set by the user for fast networks or to a second reassembly timer set by the user for slow networks based on whether the receiver and sender are on the same subnet.

Abstract: A method for enhancing port allocation procedures in a computer network by reserving specific ports to particular users and processes. Access protocol for ports are modified to enable a system administrator to defined/pre-select particular users and processes that may access the specific ports. A table structure is provided with the list of users and processes with bind authority for the particular port. When a bind request is received the ID of the user or process is confirmed against those within the table of the requested port. Bind access is provided only when the user ID or process ID matches one that is within the table. The port is allocated to the user/program until the user/process is complete, and other requests for access to the port are ignored until final completion of the ongoing process, even when a temporary disconnect of the port occurs.