Skip to content

Latest commit

 

History

History
81 lines (63 loc) · 2.62 KB

File metadata and controls

81 lines (63 loc) · 2.62 KB

OpenLogReplicator.json - metrics element

Author: Adam Leszczyński <aleszczynski@bersler.com>, version: 1.9.0, date: 2026-01-23

This section describes the metrics element in the OpenLogReplicator JSON configuration. It documents available fields, constraints, semantics and operational notes for the supported metrics backends.

Tip

Metrics are used by all OpenLogReplicator tutorials to monitor OpenLogReplicator Docker container. Refer to full list of available counters and gauges in Metrics guide.

Table 1. Metrics element
Parameter Type / constraints Description and notes

type

string, max length: 128, mandatory

Name of the metrics backend module. Currently the only supported value is prometheus.

NOTE: Unknown type values are rejected at startup.

bind

string, max length: 128, mandatory for prometheus

Network address the metrics server binds to, in the form <host>:<port>. Prometheus scrapes metrics by connecting to this address.

Example: "bind": "127.0.0.1:8080"

CAUTION: Bind address must be reachable by Prometheus and writable by the process user.

tag-names

string, max length: 128, default: none

Controls which additional label tags are attached to dml_op and related metrics.

Allowed values:

  • none — (default) do not attach schema / table labels.

  • filter — attach schema and table labels only for tables matched by the configured filter rules.

  • sys — attach schema and table labels only for system tables required by OLR.

  • all — attach schema and table labels for all emitted metrics (equivalent to filter + sys).

TIP: Use filter to limit cardinality of label values and avoid high-cardinality metrics which can harm Prometheus performance.

Note
  • Validate bind and firewall rules before enabling Prometheus scraping.

  • Prefer filter or sys over all in production to limit label cardinality.

  • When exposing metrics on public interfaces, secure access via firewall or a reverse proxy to avoid information disclosure.

  • Restart the process after changing metrics configuration to ensure the bind address and module are re-initialized.

Example metrics configuration (JSON)
{
  "metrics": {
    "type": "prometheus",
    "bind": "127.0.0.1:8080",
    "tag-names": "filter"
  }
}