Courier Fetch: 70 of 80 shards failed

[manopmk (manojkumar)]

i have added multiple log in logstash conf
if i refresh kibana4 showing this
Courier Fetch: 70 of 80 shards failed.
for this i searched in google they told to add
this line in elasticsearch.yml

threadpool.search.type: fixed
threadpool.search.size: 200
threadpool.search.queue_size: 20000

but still its showing that error

note:im using mac
using local ip

[weltenwort (Felix Stürmer)]

Hi @manopmk,

this sounds like Elasticsearch does not have enough resources to completely answer the query. Increasing the search threadpool size is a measure that might help, but like all performance parameters it is highly dependent on the runtime environment. The Elasticsearch log output might give some indication about the reason for the 10 failed shards.

[manopmk (manojkumar)]

Thank you buddy
in es log
org.elasticsearch.search.SearchParseException: [logstash-2016.10.21][4]: from[-1],size[500]: Parse Failure [Failed to parse source [{"size":500,"sort":{"tstamp":"asc"},"query":{"filtered":{"query":{"query_string":{"analyze_wildcard":true,"query":""}},"filter":{"bool":{"must":[{"range":{"tstamp":{"gte":1490595144374,"lte":1490609544374}}}],"must_not":[]}}}},"highlight":{"pre_tags":["@kibana-highlighted-field@"],"post_tags":["@/kibana-highlighted-field@"],"fields":{"":{}}},"aggs":{"2":{"date_histogram":{"field":"tstamp","interval":"5m","pre_zone":"+05:30","pre_zone_adjust_large_interval":true,"min_doc_count":0,"extended_bounds":{"min":1490595144374,"max":1490609544374}}}},"fields":["*","_source"],"script_fields":{},"fielddata_fields":["_timestamp","@timestamp","tstamp"]}]]
Parse Failure [No mapping found for [tstamp] in order to sort on]];

Actullay i created index called tstamp which contain exact time stamp of log
my conf

filter {
grok {
match => [ "message", "%{DATESTAMP:timestamp}" ]
}
date {
locale => "en"
match => [ "timestamp", "MM-dd-YYYY HH:mm:ss","ISO8601" ]
target => ["tstamp"]
remove_field => ["timestamp"]
}
}
Json i got
"message" => "03-27-2017 10:26:11 INFO - in DAO Impl query: fetching username from username_notification table",
"@version" => "1",
"@timestamp" => "2017-03-27T09:33:50.768Z",
"host" => "Vishnu-Prasad.local",
"path" => "/Users/tcstsb3/Downloads/batch/starbuck-batch-service.log",
"tstamp" => "2017-03-27T04:56:11.000Z"
}

in kibana i added new index called tstamp i overrided the changes
but i dont know why this showing No mapping found

can you plz help me