我建立了一个Redis集群，其中包含4个主节点和4个从属节点，并具有以下节点信息：
127.0.0.1:8001> cluster nodes
88d364e0b20e6ec6a471ec6f12d41b82b5f5eab5 127.0.0.1:8009@18009 master - 0 1553593755000 51 connected 0-1165 5461-6627 10923-12088
e2a19e5e4e18327661837d8ce76b40648e6ac418 127.0.0.1:8003@18003 slave 6afaefabbac4d7cccbddbb1c793a7bb72cc10099 0 1553593755414 46 connected
6afaefabbac4d7cccbddbb1c793a7bb72cc10099 127.0.0.1:8007@18007 master - 0 1553593755000 46 connected 12089-16383
3ad52aacf49349d3e1173fb0545a081a2908b048 127.0.0.1:8006@18006 master - 0 1553593755113 50 connected 1166-5460
07d56713b49b4f0273a1d1ee9ad62f58613705e4 127.0.0.1:8004@18004 slave 7c4e0d35fdd1901666d7e0ed4a2298ec73c6e209 0 1553593755000 52 connected
e8c230d88f5abebcb96719cec7d665f2a57400d8 127.0.0.1:8001@18001 myself,slave 3ad52aacf49349d3e1173fb0545a081a2908b048 0 1553593755000 49 connected
7c4e0d35fdd1901666d7e0ed4a2298ec73c6e209 127.0.0.1:8008@18008 master - 0 1553593755314 52 connected 6628-10922
78c8f2ab053b0c2ce2f0635ad042e6c20ca707eb 127.0.0.1:8002@18002 slave 88d364e0b20e6ec6a471ec6f12d41b82b5f5eab5 0 1553593755000 51 connected
127.0.0.1:8001>
我希望如果任何具有单个从属的主机立即发生故障，则应通过跳过选举过程将其升级为主机。有什么办法可以做到这一点？