Skip to content
BleemeoBleemeo

Web Interface

Free
Starter
Professional

The Bleemeo agent runs a local web interface you can reach at http://localhost:8015, which provides a quick overview of your server’s health.

Bleemeo agent local web interface showing the dashboard with CPU, memory and network metrics

This interface is read-only and listens on localhost only by default. You may disable it with the web.enable setting in the agent configuration.

The interface has no authentication. To keep other websites from reading it through your browser, the agent only answers a browser that reaches it by an IP address or by localhost. If you open it with another host name (for instance through a reverse proxy, or after changing web.listener.address), add that name to web.listener.allowed_hosts, otherwise the page answers Glouton refused this request.

Uptime Monitoring tab

Section titled “Uptime Monitoring tab”

When at least one HTTP, TCP, ICMP, DNS or SSL target is declared under blackbox.targets, the local UI exposes a dedicated Uptime Monitoring tab. Each configured monitor gets a card showing:

  • current up/down status (with a short failure reason when down: TLS error, DNS failure, HTTP status code, body mismatch, or timeout)
  • a 24-hour probe_success sparkline
  • uptime percentage over the same window
  • p95 response time over 24 hours
  • days until certificate expiry for HTTPS targets

The page also shows a KPI row (count of monitors up, average uptime, currently-down list) and a banner that lists certificates expiring within 14 days.

Clicking a card opens a drawer with the per-monitor availability stripe, a recent failures list with probable cause, response time chart, HTTP phase breakdown (resolve / connect / TLS / processing / transfer), HTTP status code timeline, and the full TLS certificate detail when applicable. The drawer URL is shareable (http://localhost:8015/monitors/<monitor-name>).

Monitors provisioned from Bleemeo Cloud are also listed here and are visually flagged with a “Bleemeo” badge to distinguish them from local-config targets.

History beyond the in-memory window (3 minutes) requires the local TSDB to be enabled — see agent.local_store.enable.

Thresholds

Section titled “Thresholds”

The local UI shows the thresholds configured on your metrics, whether they come from the agent configuration or from Bleemeo, and how each metric stands against them:

  • The CPU, memory, swap and load cards of the dashboard take the color of their status, and show for how long a threshold has been crossed (for instance WARNING for 12m).
  • Charts draw the warning and critical thresholds as dashed lines.
  • A banner at the top of the dashboard lists the thresholds currently crossed.
  • The Informations tab lists every active threshold in a table you can sort and search.

Containers tab

Section titled “Containers tab”

The Containers tab lists the containers of the server. Click a column header (state, name, image, CPU, memory, network, start date) to sort the list. Open a container and choose View charts to reach its own page (http://localhost:8015/containers/<name>), with the history of its CPU, memory, network and disk I/O usage.

Informations tab

Section titled “Informations tab”

Besides the agent and server details, the Informations tab shows:

  • the agent’s own logs, refreshed every 5 seconds, with a search field that highlights the matching lines. It is a quick way to spot errors without opening a shell on the server.
  • the running configuration, merged from the defaults, the configuration files and the environment variables. View running configuration opens it, and Download (.yaml) saves it. Values whose key contains key, secret, password or passwd are replaced by *****.

Other endpoints

Section titled “Other endpoints”

The local web server also exposes:

  • /metrics: a Prometheus compatible exporter of all metrics gathered by Bleemeo agent
  • /metrics?onlyMonitors: the raw probe_* metrics from the blackbox prober. Probes are excluded from the plain /metrics output by default because they trigger network calls; use this flag (or ?includeMonitors to mix in everything else) when scraping Glouton’s monitor data from an external Prometheus.
  • /diagnostic: a diagnostic page that can be useful when troubleshooting issues.
  • /data/config: the running configuration as YAML, with secrets redacted as described above.
  • /api/v1/query, /api/v1/query_range, /api/v1/series, /api/v1/labels and /api/v1/label/<name>/values: a Prometheus-compatible query API on the metrics of the agent. You can add the agent as a Prometheus data source in Grafana, including for stat, gauge and table panels and for template variables. Only the last 3 minutes are available unless the local TSDB is enabled.