TimescaleDB Integration
Monitor hypertable chunk metrics, compression ratios, and continuous aggregate health across your TimescaleDB deployments. Get AI-powered ingestion anomaly detection and background job failure alerts.
How It Works
Deploy TigerOps Postgres Agent
Install the TigerOps agent with the TimescaleDB extension enabled. The agent connects to Postgres and queries timescaledb_information views to collect hypertable, chunk, and compression statistics.
Configure Hypertable Monitoring
Specify which hypertables to monitor by name or schema pattern. TigerOps collects per-hypertable chunk counts, compressed chunk ratios, row counts, and disk size — both pre- and post-compression.
Enable Background Job Tracking
TigerOps monitors TimescaleDB background jobs (compression policies, data retention, continuous aggregate refresh) for execution success, duration, and failure counts per job type.
Set Chunk and Compression Alerts
Define thresholds for uncompressed chunk accumulation, continuous aggregate refresh lag, and retention policy failures. TigerOps correlates background job failures with query latency degradation.
What You Get Out of the Box
Hypertable Chunk Metrics
Per-hypertable chunk count, uncompressed vs compressed chunk ratio, oldest chunk age, and total data volume. Track ingestion rates and detect when chunk accumulation is outpacing compression policies.
Compression Ratio Monitoring
Per-hypertable compression ratios, bytes saved, compression success rates, and recompression frequency. TigerOps alerts when compression efficiency degrades or compression policies fall behind ingestion.
Continuous Aggregate Health
Continuous aggregate refresh job success rates, refresh duration, last successful refresh timestamp, and staleness lag. Alert when a continuous aggregate falls behind its refresh interval.
Data Retention Policy Tracking
Retention policy execution status, chunks dropped per run, and estimated disk reclaimed. TigerOps tracks whether retention policies are executing on schedule and alerts on failures.
Time-Series Query Performance
Postgres query latency for time-series scans, chunk exclusion efficiency (chunks skipped by time predicate pushdown), and parallel query worker utilization for large time-range queries.
AI Ingestion Anomaly Detection
TigerOps AI baselines your hypertable ingestion rate patterns and fires alerts when ingestion drops below expected volume, indicating upstream data pipeline failures before they cause data gaps.
TigerOps Agent Config for TimescaleDB
Configure the TigerOps agent to collect hypertable, compression, and background job metrics from TimescaleDB.
# TigerOps TimescaleDB integration config
# Place at /etc/tigerops/conf.d/timescaledb.yaml
integrations:
- name: timescaledb
type: timescaledb
config:
host: timescaledb.internal
port: 5432
database: tsdb
user: tigerops_monitor
password: "${TIMESCALEDB_MONITOR_PASSWORD}"
sslmode: require
# Hypertable monitoring
hypertables:
# Monitor all hypertables (or specify a list)
monitor_all: true
# schemas to include
include_schemas:
- public
- metrics
metrics:
- chunk_stats # chunk count, compression status
- compression_stats # compression ratios, bytes saved
- row_counts # estimated row counts
- disk_size # pre/post compression size
# Background job monitoring
background_jobs:
enabled: true
job_types:
- compress_chunks
- reorder_chunks
- drop_chunks # data retention
- refresh_continuous_aggregate
# Alert if job hasn't succeeded in N intervals
staleness_multiplier: 3
# Continuous aggregate monitoring
continuous_aggregates:
enabled: true
max_staleness_minutes: 60
# Standard Postgres metrics (via pg_stat_*)
postgres_metrics: true
scrape_interval: 60s # Background job stats are slow-changing
remote_write:
endpoint: https://ingest.atatus.net/api/v1/write
bearer_token: "${TIGEROPS_API_KEY}"Common Questions
Which TimescaleDB versions does TigerOps support?
TigerOps supports TimescaleDB 2.x on PostgreSQL 13, 14, 15, and 16. Timescale Cloud (managed service) is supported via the Timescale metrics API. Both single-node and multi-node TimescaleDB deployments are supported.
What Postgres permissions does the monitoring user need?
The monitoring user needs SELECT on timescaledb_information.*, SELECT on timescaledb_experimental.*, and the pg_monitor role for pg_stat_* views. For background job metrics, SELECT on _timescaledb_internal.bgw_job_stat is also needed. TigerOps provides a ready-made SQL script for creating the monitoring user.
How does TigerOps detect when compression is falling behind?
TigerOps tracks the count of uncompressed chunks older than your compression policy interval. When uncompressed chunk accumulation exceeds a configurable threshold, it fires an alert and shows the estimated disk overhead and query performance impact of the deferred compression.
Can TigerOps alert on continuous aggregate staleness?
Yes. TigerOps computes the difference between the current timestamp and the last successful continuous aggregate refresh for each aggregate. Alerts fire when staleness exceeds your configured SLO, with context on why the refresh may have failed.
Does TigerOps monitor TimescaleDB multi-node deployments?
Yes. For TimescaleDB multi-node (access node + data nodes), TigerOps monitors per-data-node chunk distribution, distributed query execution plans, and data node connectivity health. Access node query routing latency and data node skew are tracked separately.
Keep Your TimescaleDB Policies Running on Schedule
Hypertable monitoring, compression tracking, continuous aggregate health, and AI ingestion anomaly detection. Deploy in 5 minutes.