Hello,
I am currently conducting performance tests on an OpenSearch cluster running in a Kubernetes environment. The cluster is configured with three indexing nodes, and logs are collected via Fluent Bit. The purpose of these tests is to observe the system’s data processing capacity and its limits.
During the tests, the following error messages were observed in the Fluent Bit logs:
May 25 10:27:00 fluent-bit[414235]: {"error":{"root_cause":[{"type":"rejected_execution_exception","reason":"rejected execution of coordinating operation [coordinating_and_primary_bytes=536769568, replica_bytes=0, all_bytes=536769568, coordinating_operation_bytes=2656432, max_coordinating_and_primary_bytes=536870912]"}],"type":"rejected_execution_exception","reason":"rejected execution of coordinating operation [coordinating_and_primary_bytes=536769568, replica_bytes=0, all_bytes=536769568, coordinating_operation_bytes=2656432, max_coordinating_and_primary_bytes=536870912]"},"status":429}
However, the output of the _cluster/stats
API indicates that the system has a higher resource usage capacity than this (512MB) limit:
"jvm": {
"max_uptime": "10.1d",
"max_uptime_in_millis": 877134375,
"versions": [
{
"version": "21.0.6",
"vm_name": "OpenJDK 64-Bit Server VM",
"vm_version": "21.0.6+7-LTS",
"vm_vendor": "Eclipse Adoptium",
"bundled_jdk": true,
"using_bundled_jdk": true,
"count": 3
}
],
"mem": {
"heap_used": "8.8gb",
"heap_used_in_bytes": 9554800536,
"heap_max": "15gb",
"heap_max_in_bytes": 16106127360
},
"threads": 282
}
I have no idea how to increase this limit. Could you please help me?
Also my fluent-bit.conf file is like below
[SERVICE]
Flush 1
Log_Level info
Parsers_File parsers.conf
[INPUT]
Name dummy
Rate 2500
Copies 10
Dummy {"message":"25k EPS test log"}
[OUTPUT]
Name opensearch
Match *
Host fqdn
Port 443
Index test-25k
Type _doc
HTTP_User admin
HTTP_Passwd pass
tls On
Logstash_Format Off
Suppress_Type_Name On