Description
The API request Rate limits are 25 Write requests per second or 75 Read requests per second, across all API keys pointing to a particular LSRC. If the number of requests go beyond the above mentioned limits, you may see the requests being delayed, throttled or rejected.
This article outlines the Metrics API commands, to monitor the read and write requests for a Confluent Cloud Schema Registry cluster.
Applies To
Confluent Cloud
Schema Registry
Resolution
To monitor the requests rate at a particular time, you can query the Metrics API similar to these commands.
For Write requests:
curl -d '{
"aggregations": [
{
"metric": "io.confluent.kafka.schema_registry/schema_operations_count",
"agg": "SUM"
}
],
"filter": {
"filters": [
{
"field": "resource.schema_registry.id",
"op": "EQ",
"value": "lsrc-"
},
{
"field": "metric.method",
"op": "EQ",
"value": "CREATE"
}
],
"op": "AND"
},
"granularity": "PT1M",
"intervals": [
"now-5m/now"
],
"group_by": [
"resource.schema_registry.id"
]
} ' --header 'content-type: application/json' \
"https://api.telemetry.confluent.cloud/v2/metrics/cloud/query" \
-u key:secret | jq
For Read requests:
curl -d '{
"aggregations": [
{
"metric": "io.confluent.kafka.schema_registry/schema_operations_count",
"agg": "SUM"
}
],
"filter": {
"filters": [
{
"field": "resource.schema_registry.id",
"op": "EQ",
"value": "lsrc-"
},
{
"field": "metric.method",
"op": "EQ",
"value": "READ"
}
],
"op": "AND"
},
"granularity": "PT1M",
"intervals": [
"now-5m/now"
],
"group_by": [
"resource.schema_registry.id"
]
} ' --header 'content-type: application/json' \
"https://api.telemetry.confluent.cloud/v2/metrics/cloud/query" \
-u key:secret | jqFrom the above queries, please replace lsrc- with the complete lsrc id details and choose the intervals accordingly.