We need to test the error propagation behavior when range deletion fails due to ingress operation rate limiter rejection. We should cover both the budget exhaustion and retry cap scenarios.

Given the range deleter runs in an AlternativeClientRegion context with no transport session, isProcessInternalClient()returns true and it is unconditionally exempt from rate limiting. The test asserts that orphaned documents are cleaned up after a moveChunk even with the rate limiter maximally overloaded (defaultClientMaxRetryAttempts: 0, near-zero token rate), confirming both range-deletion sub-paths are covered:

• Local orphan deletes: range-deleter-batch client with no transport session → isProcessInternalClient() = true → exempt
• Inter-shard coordination: Sharding-Fixed → exempt by app name

This approach was confirmed by Vojislav. See also SERVER-127637 for the analogous chunk migration exemption test, and SERVER-127632 for the reference pattern.