High disk watermark Error while Running Elasticsearch

Getting the following error while running elastic search.
ussjvlx1128 [6:43] [/home/xvivech] -> [2016-08-17 06:44:26,762][WARN ][cluster.routing.allocation.decider] [Clea] high disk watermark [90%] exceeded on [ZlnSBDYcQPCxB_RnggHtBg][Clea][/home/xvivech/log_mgt/elasticsearch-2.0.0/data/elasticsearch/nodes/0] free: 507.8mb[9.9%], shards will be relocated away from this node

The elasticsearch API were working fine earlier but now getting the follow error

TRACE: 2016-08-17T12:42:16Z
"file": "getchassis",
"params": {
"log_type": true
<- 503
"error": {
"root_cause": [],
"type": "search_phase_execution_exception",
"reason": "all shards failed",
"phase": "query",
"grouped": true,
"failed_shards": []
"status": 503

I'd suggest that you have unallocated shards due to a full disk.

I seem to be having a similar issue. Win2003 server that elastic was running on ran low on space, and due to a crash had to be forcefully restarted.

I have a cluster status of RED now with 9 unassigned shards.
radiology 4 p STARTED 434927 502.5mb Wolf
radiology 4 r UNASSIGNED
radiology 0 p STARTED 434681 503.6mb Wolf
radiology 0 r UNASSIGNED
radiology 3 p UNASSIGNED
radiology 3 r UNASSIGNED
radiology 1 p UNASSIGNED
radiology 1 r UNASSIGNED
radiology 2 p STARTED 434642 505.6mb Wolf
radiology 2 r UNASSIGNED
.kibana 0 p UNASSIGNED
.kibana 0 r UNASSIGNED

I'm trying to assign these shards so the cluster will run again, but I'm struggling badly.
The reroute -d command runs, but never actaully moved a shard into the running state.