You can configure load shedding rules to determine how an MRA device reacts to a processing backlog. (Refer to Configuring Load Shedding for an MRA Device.) This state is called “busyness.” Levels of busyness can be configured to accept, reject, or drop select messages at each level. An MRA has two busyness levels. At any level of busyness, request that have been queued longer than a configurable time are silently discarded without further processing, since the originator would have already given up on that request.
Level actions are configurable. The Table 1 and Table 2 tables show the default load-shedding rules for an MRA device.
The default action for Level 1 is Accept.
Rule Name | Actions |
---|---|
DefaultRule1 | Reject Gx CCR messages with DIAMETER_TOO_BUSY |
DefaultRule2 | Reject Gxx CCR messages with DIAMETER_TOO_BUSY |
Rule Name | Actions |
---|---|
DefaultRule3 | Accept Drma LNR with ACCEPT |
DefaultRule4 | Accept Drma LNR with ACCEPT |
DefaultRule5 | Accept Drma LNR with ACCEPT |