wonky explains

> db.data_points.find({ sensor_id: 247, faulty: { $ne: true }, orig_relative_time: { $lte: new Date(1299305991280) }, deleted_at: { $exists: false } }).sort({ orig_relative_time: -1 }).limit(-1).explain()
{
	"cursor" : "BtreeCursor orig_relative_time_-1_sensor_id_-1",
	"nscanned" : 11859,
	"nscannedObjects" : 11859,
	"n" : 11859,
	"millis" : 1896,
	"indexBounds" : {
		"orig_relative_time" : [
			[
				"Sat Mar 05 2011 06:19:51 GMT+0000 (UTC)",
				"Thu Jan 01 1970 00:00:00 GMT+0000 (UTC)"
			]
		],
		"sensor_id" : [
			[
				247,
				247
			]
		]
	}
}



> db.data_points.find({ sensor_id: 247, faulty: { $ne: true }, orig_relative_time: { $lte: new Date(1299305991280) }, deleted_at: { $exists: false } }).sort({ orig_relative_time: -1 }).limit(1).explain() 
{
	"cursor" : "BtreeCursor sensor_id_-1_orig_relative_time_-1",
	"nscanned" : 1,
	"nscannedObjects" : 1,
	"n" : 1,
	"millis" : 0,
	"indexBounds" : {
		"sensor_id" : [
			[
				247,
				247
			]
		],
		"orig_relative_time" : [
			[
  			"Sat Mar 05 2011 06:19:51 GMT+0000 (UTC)",
				"Thu Jan 01 1970 00:00:00 GMT+0000 (UTC)"
			]
		]
	}
}

I should note that the millis in the first explain is very high...the DB was getting hammered then so I suspect that is what caused it to be slow. After it warmed up it settled down, but still fairly high, in the 80-150 range