Failed Short Message Delivery

When the VMSC/SGSN indicates a short message delivery failure, the SMSC may send a message to the HLR, using the MAP_REPORT_SM_DELIVERY_STATUS procedure, indicating the reason for the delivery failure and requesting that the SMSC be put on a list of service centres wanting to be notified when the destination party becomes available again. The HLR will set a flag against the destination account, indicating that it is unavailable for short message delivery, and store the SMSC's address in the Message Waiting Delivery (MWD) list for the destination party. Valid flags are Mobile Not Reachable Flag (MNRF), Memory Capacity Exceeded Flag (MCEF) and Mobile Not Reachable for GPRS (MNRG). The HLR will now start responding to SRI-for-SM requests with a failure, indicating the failure reason, and will automatically add the requesting SMSC's address to the MWD list for the destination party.

The HLR may be informed of a subscriber becoming available for short message delivery in several ways:

* Where the subscriber has been detached from the network, a reattach will trigger a Location Update to the HLR.
* Where the subscriber has been out of coverage, but not fully detached from the network, on coming back into coverage it will respond to page requests from the Visitor Location Register (VLR). The VLR will then send a Ready-for-SM (mobile present) message to the HLR.
* Where the MS has had its memory full, and the subscriber deletes some texts, a Ready-for-SM (memory available) message is sent from the VMSC/VLR to the HLR.

Upon receipt of an indication that the destination party is now ready to receive short messages, the HLR sends an AlertSC MAP message to each of the SMSCs registered in the MWD list for the subscriber, causing the SMSC to start the Short Message delivery process again, from the beginning.[1]

Additionally, the SMSC will go into a retry schedule, attempting to periodically deliver the SM without getting an alert. The retry schedule interval will depend on the original failure cause - transient network failures will result in short retry schedule, whereas out of coverage will typically result in a longer schedule.