Public Metrics
Metrics play a crucial role in monitoring and analyzing the performance of Oxla, enabling users to gain valuable insights into their data processing and analytics workflows. The metrics are specific to a single node.
This documentation will guide you on how to access and leverage the metrics.
All metrics exposed by Oxla have a label_node_id with a node ID set. This is the same node_id that can be found in logs and on data storage.
Once you have successfully deployed Oxla, you can access the comprehensive metrics that provide valuable insights into the performance and health of your database. To access the metrics, follow the steps below:
1) Open your preferred web browser and enter the following URL:
Replace <ip> with the actual IP address of your Oxla deployment. For example, if your Oxla instance is hosted on IP address 128.128.1.1, the URL would be:
2) If you are running Oxla locally on your machine, you can use localhost as the IP address. In that case, the URL would be:
3) Upon accessing the URL, you will be presented with various metrics as shown below:
The following is a list of parameters available in the Public Metrics:
Parameters | Description |
exposer_prefix | The component responsible for exposing metrics. |
exposer_transferred_bytes_total | Total size of bytes transferred out by the endpoint. |
exposer_scrapes_total | Number of times metrics were scraped. |
exposer_request_latencies | Latencies of serving scrape requests, in microseconds. |
exposer_request_latencies_count | Count of latencies. |
exposer_request_latencies_sum | Sum of latencies. |
client_sql_successful_queries_total | Number of successful queries received from clients. The available queries are:
The available label is query_type. |
client_sql_queries_total | Number of queries received from clients. The available queries are:
The available label is query_type. |
file_flushed_total | Number of files flushed when inserting rows. |
tasks_executed_total | Number of tasks executed on the cluster. |
tasks_received_total | Number of tasks a node has received to be executed. The available task types are:
|
tasks_scheduled_total | Number of tasks the leader has sent to nodes and are being executed. The available task types are:
|
thread_pool_tasks_finished_total | Number of tasks finished by the thread pool. |
thread_pool_tasks_started_total | Number of tasks started by the thread pool. |
readers_closed_total | Total number of readers closed since Oxla process creation. |
readers_opened_total | Total number of readers opened since Oxla process creation. |
writers_closed_total | Total number of writers closed since Oxla process creation. |
writers_opened_total | Total number of writers opened since Oxla process creation. |
cluster_has_leader_bool | Indicates if the cluster has a leader or not. |
num_nodes_connected | Number of nodes connected. |
node_is_leader_bool | Indicates if the node is a leader. |
num_open_connections | Number of open connections. |
postgres_connections_total | Number of clients connected using the Postgres protocol. |
tasks_capacity_total | Number of task capacity on the cluster. |
tasks_ongoing_total | Number of tasks currently being executed on the node. The available task types are:
|
thread_pool_size_total | Number of threads used by the thread pool. |
oxla_memory_usage_bytes | Memory used by Oxla. The available values are:
|
oxla_mallinfo | Information about glibc's malloc() internal state. |
ā¹ļø By regularly monitoring these metrics, you can gain valuable insights into the efficiency of your database operations, identify potential bottlenecks, and make data-driven optimizations to enhance the overall performance and reliability of your analytical workflows.
ļ»æ
