tutorial

Monitoring Secoda with Vigilmon: AI Data Catalog Health, API Availability & Search Service Uptime

How to monitor Secoda AI-powered data catalog infrastructure with Vigilmon — API availability, dashboard uptime, search service health, and SSL certificate validity.

Secoda is the AI-powered data catalog and knowledge hub that centralizes table documentation, lineage, ownership, and usage metrics into a single platform, and lets data teams ask questions in natural language to find the data they need. When Secoda's API goes down, the search and discovery workflows that data analysts depend on are blocked, new documentation is not indexed, and the Slack and VS Code integrations that surface Secoda context inline go silent. When the dashboard becomes unavailable, the institutional knowledge that Secoda has accumulated — ownership assignments, business glossary definitions, and data lineage graphs — becomes inaccessible at exactly the moment engineers need it during a data incident. Vigilmon gives you external visibility into Secoda's platform availability: the API, dashboard, AI search service, and webhook integrations, so you know the moment the knowledge hub itself goes dark.

What You'll Build

  • An HTTP monitor on the Secoda dashboard to catch frontend failures
  • A health-check monitor on the Secoda REST API to detect backend degradation
  • A monitor on the Secoda search service to verify AI-powered queries are responding
  • An SSL certificate monitor for your Secoda domain
  • Alerting configured with thresholds appropriate to each component

Prerequisites

  • A Secoda account (cloud-hosted SaaS or self-hosted deployment)
  • The Secoda dashboard accessible via HTTPS (e.g., https://app.secoda.co or your workspace subdomain)
  • API token generated in Secoda's workspace settings
  • A free account at vigilmon.online

Step 1: Understand Secoda's Service Architecture

Secoda's platform is composed of interconnected services:

| Component | Access point | Role | |---|---|---| | Secoda Dashboard | HTTPS (web UI) | Catalog browsing, documentation editing, lineage exploration | | Secoda REST API | HTTPS | Programmatic catalog queries, metadata retrieval, search | | AI search engine | Internal | Powers natural language queries across catalog assets | | Indexing service | Internal | Syncs metadata from connected warehouses, BI tools, and dbt | | Integration webhooks | Outbound | Delivers notifications to Slack, MS Teams, and custom tools |

External monitoring focuses on the API and dashboard HTTPS endpoints — these are the interfaces your team and integrations interact with, and their availability determines whether Secoda's value is accessible.

Step 2: Monitor the Secoda Dashboard

The Secoda dashboard is where data consumers search for data, data stewards document assets, and engineers trace lineage during incidents:

curl -I https://app.secoda.co
# Returns HTTP 200 with HTML when the dashboard is available
  1. Log in to Vigilmon → Add Monitor → HTTP.
  2. URL: https://app.secoda.co (or your workspace-specific URL).
  3. Check interval: 60 seconds.
  4. Response timeout: 15 seconds.
  5. Expected status: 200.
  6. Keyword: Secoda (appears in page title and HTML content).
  7. Click Save.

This monitor catches:

  • Secoda frontend process crashes or failed deployments
  • CDN or reverse proxy misconfigurations
  • DNS failures for the Secoda workspace domain
  • Failed product updates that leave the frontend broken

Alert sensitivity: 1 consecutive failure. Secoda's dashboard is where data consumers go first when they need to understand data — unavailability during business hours interrupts data discovery, documentation updates, and incident investigation workflows across the entire data organization.

Step 3: Monitor the Secoda REST API

Secoda's API powers programmatic integrations — dbt CI/CD pipelines that look up table ownership before raising a pull request, custom data portals, and VS Code extensions that surface Secoda context inline:

curl -H "Authorization: Bearer <api_token>" \
  "https://api.secoda.co/v1/health"
# Returns: {"status": "healthy"} when the API is responsive
  1. Add Monitor → HTTP.
  2. URL: https://api.secoda.co/v1/health.
  3. Headers: Authorization: Bearer <your-api-token>.
  4. Check interval: 60 seconds.
  5. Response timeout: 10 seconds.
  6. Expected status: 200.
  7. Keyword: healthy.
  8. Label: Secoda API health.
  9. Click Save.

Why monitor the API separately? The Secoda dashboard may appear to load from CDN cache even when the API backend is down. A failing API means search queries, lineage lookups, and metadata updates are all blocked — including the VS Code and Slack integrations that data engineers use inline without opening the browser.

Step 4: Monitor the Secoda Search Service

Secoda's AI-powered natural language search is the primary discovery interface for many data teams. When the search backend is degraded, queries return empty results or errors even when the dashboard HTML loads:

curl -H "Authorization: Bearer <api_token>" \
  "https://api.secoda.co/v1/search?q=revenue&limit=1"
# Returns search results when the search service is healthy
  1. Add Monitor → HTTP.
  2. URL: https://api.secoda.co/v1/search?q=revenue&limit=1.
  3. Headers: Authorization: Bearer <your-api-token>.
  4. Check interval: 3 minutes.
  5. Response timeout: 20 seconds.
  6. Expected status: 200.
  7. Keyword: results or count (present in search response JSON).
  8. Label: Secoda search service.
  9. Click Save.

This catches degradation in the AI search index or the natural language query engine — a subset of failures that a general API health check would miss. Search timeouts longer than 20 seconds indicate the AI engine is overloaded or the search index is being rebuilt.

Step 5: Monitor the Secoda Metadata Sync Endpoint

Secoda pulls metadata from connected sources (Snowflake, BigQuery, dbt, Looker, Tableau) on a schedule. A status endpoint confirms that syncs are completing and the catalog reflects recent changes:

curl -H "Authorization: Bearer <api_token>" \
  "https://api.secoda.co/v1/integrations/status"
# Returns integration sync status and last successful run timestamps
  1. Add Monitor → HTTP.
  2. URL: https://api.secoda.co/v1/integrations/status.
  3. Headers: Authorization: Bearer <your-api-token>.
  4. Check interval: 10 minutes.
  5. Response timeout: 15 seconds.
  6. Expected status: 200.
  7. Keyword: last_sync or status (present in integration status responses).
  8. Label: Secoda integration sync.
  9. Click Save.

Stale integrations mean new tables, schema changes, and freshly documented assets are not reflected in the catalog — analysts searching for recently added data assets will get empty or outdated results.

Step 6: Monitor SSL Certificates

Secoda's API is accessed by programmatic clients — VS Code extensions, CI/CD scripts, and custom data portals — that break silently when TLS certificates expire:

openssl s_client -connect api.secoda.co:443 2>/dev/null | openssl x509 -noout -dates
  1. Add Monitor → SSL Certificate.
  2. Domain: api.secoda.co.
  3. Alert when expiry is within: 30 days.
  4. Alert again: 14 days, 7 days, 3 days, 1 day.
  5. Click Save.

Also add a second SSL monitor for the dashboard domain:

  1. Add Monitor → SSL Certificate.
  2. Domain: app.secoda.co.
  3. Alert when expiry is within: 30 days.
  4. Alert again: 14 days, 7 days, 3 days, 1 day.
  5. Label: Secoda dashboard SSL.
  6. Click Save.

Two domains, two certificates: Secoda's app domain and API domain may use separate TLS certificates. A certificate expiry on the API domain breaks integrations and programmatic clients while the dashboard continues to work — and vice versa. Monitor both independently to catch either failure.

Step 7: Configure Alerting

In Vigilmon under Settings → Notifications, configure your alert channels:

| Monitor | Trigger | Action | |---|---|---| | Dashboard | Non-200 or Secoda missing | Check frontend deployment; inspect CDN/proxy configuration | | API health | Non-200 or healthy missing | Backend API down; check Secoda status; all integrations affected | | Search service | Non-200 or results missing | AI search degraded; users cannot find data assets; check search index | | Integration sync | Non-200 or last_sync missing | Catalog going stale; check warehouse and BI tool connectivity | | SSL certificates | < 30 days to expiry | Renew certificate; verify browser and API clients reconnect |

Alert after: 1 consecutive failure for dashboard and API health monitors. 2 consecutive failures for search and integration sync monitors to reduce noise from transient query timeouts and sync delays.

Common Secoda Failure Modes and What Vigilmon Catches

| Scenario | Vigilmon monitor | |---|---| | Secoda API process crash | API health monitor fires; all integrations and VS Code extension break | | Dashboard frontend failure | Dashboard monitor fires; catalog browsing and documentation blocked | | AI search index rebuild | Search monitor fires; natural language queries return empty results | | Integration sync failure | Sync monitor fires; catalog becomes stale; new tables not discoverable | | TLS certificate expired on API domain | SSL monitor alerts; integrations throw TLS errors | | TLS certificate expired on app domain | SSL monitor alerts; browser access fails | | DNS misconfiguration | All monitors fire simultaneously | | Secoda upgrade breaks API | API health monitor fires after failed deployment |

Keeping the Knowledge Hub Healthy (Beyond Infrastructure)

Vigilmon monitors Secoda's infrastructure — the API, dashboard, and core services. Catalog quality and completeness require ongoing attention within Secoda:

  • Documentation coverage: Track what percentage of your high-usage tables have descriptions, owners, and business glossary tags assigned. Infrastructure availability does not guarantee catalog quality.
  • Verification workflows: Secoda supports marking assets as verified — audit which high-traffic tables have stale or unverified documentation.
  • Access control audits: Periodically review which Secoda workspaces and collections are visible to which team members, especially as headcount grows.
  • AI assistant usage: Secoda's AI assistant quality depends on the richness of underlying documentation — invest in documentation alongside infrastructure monitoring.

Vigilmon catches platform failures. Catalog completeness, accuracy, and organizational adoption require ongoing governance within Secoda.

Secoda becomes the authoritative source of truth for your data organization only when it is always available — the moment it goes down, teams fall back to asking colleagues on Slack, writing their own documentation in Notion, and losing the institutional knowledge that Secoda is supposed to preserve. Vigilmon gives you external visibility into Secoda's availability that doesn't depend on the platform self-reporting: dashboard uptime, API health, AI search service status, integration sync freshness, and SSL certificate validity, so you know immediately when the knowledge hub needs attention and can restore access before your team reverts to workarounds.

Start monitoring Secoda in under 5 minutes — register free at vigilmon.online.

Monitor your app with Vigilmon

Free plan — 5 monitors, no credit card required. Up and running in 60 seconds.

Start free →
← Back to all articles