Redis Cluster does not respond to "get" commands when slot coverage is required and one shard goes down

I created a 3 shard cluster with 1 replica per shard using "--cluster-enabled yes --cluster-require-full-coverage yes". I then took down 1 shard completely. Healthy shards rejected "get" commands.

127.0.0.1:6380> cluster keyslot 1
(integer) 9842
127.0.0.1:6380> cluster nodes
ce7a9adfcfbe86bb22a6ebaacde49ad91ce43f83 127.0.0.1:6380 myself,master - 0 0 0 connected 5001-10000
c2433c30bd93305835e044e80b82e256efcc72dc 127.0.0.1:6379 master,fail - 1454024122191 1454024120860 3 disconnected 0-5000
11ac5917812fe7bbb6a0e49ebdf407bf58a6a043 127.0.0.1:7381 slave 5e69a6eaf79af86d4db106b6f445165d00cdac7e 0 1454026674517 7 connected
5e69a6eaf79af86d4db106b6f445165d00cdac7e 127.0.0.1:6381 master - 0 1454026677601 4 connected 10001-16383
7bcca1f09a2b901d67c788393c2ee660aa1033f3 127.0.0.1:7379 slave,fail c2433c30bd93305835e044e80b82e256efcc72dc 1454024122191 1454024121881 6 disconnected
01b2b5d5d3dcc866eeefa223acc421cf6b46b8a4 127.0.0.1:7380 slave ce7a9adfcfbe86bb22a6ebaacde49ad91ce43f83 0 1454026676570 5 connected
127.0.0.1:6380> get 1
(error) CLUSTERDOWN The cluster is down
127.0.0.1:6380> 

Comment From: madolson

Hi Parth! I fixed your issue awhile ago with allow_reads_while_cluster_down, enjoy!