Good day,
We're currently experiencing an issue with one of our index (let's call it sample_index) which has 21 primary shards with 1 replica. The primary shards of sample_index are distributed on all our nodes. Unfortunately, one of our node's disk crashed yesterday, and needs to be replaced. This made the primary shard of sample_index on that crashed node to be an unassigned primary shard, probably because it failed to be relocated due to the crash. This is causing our cluster in red state. I checked all other nodes, i can still see a primary shard of sample_index intact.
How can I delete the unassigned primary shard w/o affecting the primay shard of sample_index on other working nodes?
Here's the result when I ran allocation explain:
curl -XGET http://localhost:9200/_cluster/allocation/explain?pretty
"index" : "sample_index",
"shard" : 5,
"primary" : true,
"current_state" : "unassigned",
"unassigned_info" : {
"reason" : "NODE_LEFT",
"at" : "2024-03-14T13:30:11.936Z",
"details" : "node_left [HcOL6RzKQRegH4vwLg60jg]",
"last_allocation_status" : "no_valid_shard_copy"
},
"can_allocate" : "no_valid_shard_copy",
"allocate_explanation" : "cannot allocate because all found copies of the shard are either stale or corrupt",
"node_allocation_decisions" : [
{
"node_id" : "DugpqTy2T2mWAGCaj24gsw",
"node_name" : "inca-platform-es-nodes-001",
"transport_address" : "10.3.0.101:9300",
"node_attributes" : {
"xpack.installed" : "true",
"transform.node" : "false"
},
"node_decision" : "no",
"store" : {
"found" : false
}
}