How to optimize elasticsearch for high resin speed after reconnecting one of the nodes?

N

nikolayvaganov2017-01-19 10:00:53

elasticsearch

nikolayvaganov, 2017-01-19 10:00:53

There are about 85,000,000 documents and 30GB of data in the elastic. After restarting the server with one of the nodes, a resync started, which loaded all the nodes with some exorbitant timeout numbers (there should be five nodes in total, but the fifth one is not included in the cluster now due to the response timeout of the remaining nodes) :
{
"cluster_name" : " Prod",
"status" : "red",
"timed_out" : false,
"number_of_nodes" : 4,
"number_of_data_nodes" : 4,
"active_primary_shards" : 1863,
"active_shards" : 3069,
"relocating_shards" : 0,
"initializing_shards" : 20,
"unassigned_shards" : 6426,
"
"number_of_in_flight_fetch" : 0,
"task_max_waiting_in_queue_millis" : 3558787,
"active_shards_percent_as_number" : 32.25433526011561
}
how can i optimize this process ?

Reply

Answer the question

In order to leave comments, you need to log in

1 answer(s)

N

nikolayvaganov, 2017-01-25
@nikolayvaganov

The problem was solved by allocating more RAM for java under elastic