InfluxDB Integration
Monitor write throughput, series cardinality, query latency, and shard health across your InfluxDB deployments. Catch cardinality explosions before they exhaust memory.
How It Works
Enable InfluxDB Telemetry
Add the TigerOps output plugin to your influxdb.conf. The exporter connects to the InfluxDB HTTP API and collects internal metrics exposed at the /metrics endpoint — no separate agent required.
Configure Output Plugin
Set your TigerOps ingest endpoint and API key in the output stanza. TigerOps auto-discovers all databases, retention policies, and continuous queries in your InfluxDB instance.
Define Cardinality Alerts
Set thresholds for series cardinality per measurement and database. TigerOps uses AI to forecast cardinality explosions based on tag growth rate before they exhaust memory.
Correlate With Application Traces
TigerOps links InfluxDB write failures or query latency spikes to the application services that triggered them, giving you end-to-end context from time-series writes to slow dashboards.
What You Get Out of the Box
Write Throughput & Batching
Points per second, batch sizes, write error rates, and HTTP write latency percentiles. Identify when write pressure exceeds shard capacity before data loss occurs.
Series Cardinality Tracking
Per-measurement and per-database series counts with AI-powered trend forecasting. TigerOps alerts on cardinality explosions caused by unbounded tag values.
Query Latency Percentiles
P50, P95, and P99 query latency by database and measurement. Identify slow continuous queries, expensive GROUP BY tag operations, and suboptimal retention policies.
Shard & Storage Health
Shard group lifecycle metrics, disk usage per retention policy, and WAL size tracking. TigerOps alerts when storage growth outpaces provisioned capacity.
Compaction & WAL Metrics
TSM compaction duration, level counts, and WAL flush rates. Detect when compaction falls behind write load and impacts query performance.
Retention Policy Compliance
Monitor data expiry by retention policy, track CQ execution latency, and verify downsampling pipelines are keeping up with incoming write volume.
influxdb.conf Output Config
Configure InfluxDB to forward internal metrics to TigerOps via the Prometheus remote write or Telegraf output plugin.
# influxdb.conf — enable internal metrics endpoint
[http]
bind-address = ":8086"
# Expose /metrics for Prometheus scraping
pprof-enabled = true
[monitor]
# Write internal stats every 10s to _internal database
store-enabled = true
store-database = "_internal"
store-interval = "10s"
---
# telegraf.conf — collect InfluxDB internals and forward to TigerOps
[[inputs.influxdb]]
urls = ["http://localhost:8086/debug/vars"]
[[inputs.prometheus]]
urls = ["http://localhost:8086/metrics"]
metric_version = 2
[[outputs.http]]
url = "https://ingest.atatus.net/api/v1/write"
data_format = "prometheusremotewrite"
[outputs.http.headers]
"X-TigerOps-API-Key" = "${TIGEROPS_API_KEY}"
"Content-Type" = "application/x-protobuf"
"Content-Encoding" = "snappy"
# InfluxDB 2.x — scrape via the /metrics endpoint directly
# Set INFLUXDB_TOKEN in your environment for authenticated access
[[inputs.prometheus]]
urls = ["http://localhost:8086/metrics"]
bearer_token = "${INFLUXDB_TOKEN}"
metric_version = 2Common Questions
Does TigerOps support InfluxDB OSS, InfluxDB Cloud, and InfluxDB Clustered?
Yes. TigerOps supports InfluxDB OSS 1.x and 2.x, InfluxDB Cloud (multi-tenant and dedicated), and InfluxDB Clustered. The exporter automatically adapts its metric collection based on the API version detected.
How does TigerOps detect cardinality explosions?
TigerOps continuously tracks series counts per measurement and applies time-series forecasting to project cardinality growth. It fires pre-emptive alerts when a measurement is on track to exceed your configured limit within a configurable look-ahead window.
Can I monitor InfluxDB running on Kubernetes?
Yes. The TigerOps InfluxDB exporter runs as a sidecar or Deployment alongside your InfluxDB StatefulSet. It uses the Kubernetes service DNS to discover the InfluxDB endpoint automatically.
Does TigerOps support Flux and InfluxQL query monitoring?
Yes. TigerOps captures query durations and error rates for both the Flux and InfluxQL query engines. You can filter dashboards by language and identify which query type is contributing to latency.
Can TigerOps alert on high WAL size before a crash?
Yes. TigerOps tracks the WAL size in bytes and the number of segments awaiting compaction. Alerts fire when the WAL grows beyond your configured threshold, which typically precedes write stalls or OOM conditions.
Full Visibility Into Your Time-Series Layer
No credit card required. Connect in minutes. Cardinality, write throughput, and query latency immediately.