End-to-End Monitoring: Dashboard
Overview
Use this page to check real-time end-to-end monitoring health across environments and agents. The End-to-End Monitoring Dashboard is the live cockpit for synthetic session monitoring: it aggregates session counts, success rate, errors, captured measurements, and LoadGen Agent status so you can tell whether your monitored flows are passing or failing without opening each run individually.
Before this page has data, a monitoring profile must exist and run (End-to-End Monitoring: Profiles) on an enabled LoadGen Agent (Configuration: End-to-End Monitoring: Platform Agents) against a connected datasource (Configuration: Datasources), usually on a schedule (End-to-End Monitoring: Schedules). Session results land here first; deeper analysis moves to Analyse: Dashboards and Analyse: Analysis, and the resulting health feeds Service Levels: Dashboard.
On this page, you can filter by datasource, profile, test, and measurement, switch between short and wide time windows, toggle automatic refresh, and switch the Trends chart between Mean, P95, and Success % metrics. The LoadGen Agents panel shows which agents are online, delayed, reporting no data, or disabled, and a banner flags credentials nearing expiry before they can break a monitored flow.
Key concepts
- Datasource - The monitoring data source the dashboard queries for session, trend, and agent data. (related: Configuration: Datasources)
- Profile - An End-to-End Monitoring profile defines a monitored flow (a set of tests run against a target) that this dashboard reports on. (related: End-to-End Monitoring: Profiles)
- LoadGen Agent - The process that executes monitored sessions and reports their results back to this dashboard. (related: Configuration: End-to-End Monitoring: Platform Agents)
- Measurement - A captured metric or step within a monitored session, such as a timed action in a flow, that feeds the KPI cards and Trends chart.
- Credential Profile - A stored authentication credential (such as an API token) used by monitored flows; the dashboard warns here when one is close to expiring. (related: Configuration: Credential Profiles)
When to use it
- Check whether monitored flows are currently passing or failing before escalating an incident.
- Investigate a sudden drop in success rate or a spike in errors across environments.
- Confirm a LoadGen Agent is online before or after a scheduled monitoring run.
- Compare current response-time trends (Mean or P95) against recent history.
- Catch a credential that is about to expire before it breaks a monitored flow.
- Narrow results to a specific profile, test, or measurement while troubleshooting a single flow.
Before you begin
- At least one End-to-End Monitoring profile exists and has run (created on End-to-End Monitoring: Profiles).
- A LoadGen Agent is configured and enabled (Configuration: End-to-End Monitoring: Platform Agents) so sessions can execute and report status.
- A monitoring datasource is connected (Configuration: Datasources) for the dashboard to query.
- Any credentials used by monitored flows are valid and not expired (Configuration: Credential Profiles), or this page will show a Credentials expiring warning.
Common workflows
Check current end-to-end monitoring health
Persona: Load-test operator or on-call admin
- Open the End-to-End Monitoring Dashboard and stay on the Cockpit tab.
- Expect: The Sessions, Success Rate, Errors, Measurements, and Data Freshness KPI cards load for the default 1h window with Auto refresh on.
- Review the LoadGen Agents panel.
- Expect: The Online, Delayed, No Data, and Disabled counts and the per-agent table show which agents are currently reporting.
- Select the 24h or 7d time range.
- Expect: The KPI cards and Trends chart reload for the wider window; this is read-only, no data is changed.
Verify: Success Rate holds at or near 100% and Errors stay at 0 for the selected window; if Data Freshness still reads No data, no session has reported in that window yet.
Narrow the dashboard to one monitored profile
Persona: Operator troubleshooting a specific flow
- Open the Profiles dropdown and select the profile you want to inspect.
- Expect: The Profiles button label updates to show the number selected and the page re-queries.
- Open the Tests dropdown and select a specific test within that profile.
- Expect: The Tests button updates and the data narrows further to that test.
- Read the KPI cards and Trends chart.
- Expect: They now reflect only sessions from the selected profile and test.
Verify: The Profiles and Tests buttons show a selected count instead of All, and the KPI numbers change from the unfiltered view; if the numbers look unchanged, confirm the profile ran sessions in the current time range.
Resolve an expiring monitoring credential
Persona: Admin maintaining monitored flows
- Read the Credentials expiring banner.
- Expect: It names the specific token and shows how many days remain before it expires.
- Open Configuration → Credential Profiles as the banner directs.
- Expect: You leave the dashboard for the Credential Profiles list, where the token can be reissued.
- Dismiss the banner with its close icon after reissuing the credential.
- Expect: The banner disappears from the current view.
Verify: The Credentials expiring banner no longer references that token on the next dashboard load; if it still does, confirm the new credential was saved on Credential Profiles.
Steps
Step 1
Open the End-to-End Monitoring Dashboard.
Expected result: The Cockpit tab loads with the 1h time range selected: the five KPI cards (Sessions, Success Rate, Errors, Measurements, Data Freshness) read zero or No data, the Trends chart shows no data for the window, and the LoadGen Agents panel lists one agent as Disabled with no Last Run. A Credentials expiring banner also appears, warning that an Orders API token expires in 4 days.
This is the starting view for judging whether monitored flows are currently healthy before you filter or drill into specific profiles, tests, or agents.
- The Cockpit and Analytics tabs sit below the filter bar; Cockpit is selected by default.
- Auto is checked with a 30s refresh interval, so the page re-queries on its own.
- The Credentials expiring banner can be closed with its X without resolving the underlying token.
Tabs on this page
-
Cockpit - The default real-time view: KPI cards, the Trends chart, the LoadGen Agents panel, and a paged session list with per-session AI analysis.
- Sessions, Success Rate, Errors, Measurements, Data Freshness KPI cards; Trends chart (Mean/P95/Success %); LoadGen Agents status panel and table; paged session list.
-
Analytics - Not captured in this run; switches away from the live Cockpit view toward an analytics-oriented look at the same monitoring data.
- Not captured in this run.
Field reference
| Field | Purpose | Format / Validation | Required | Default |
|---|---|---|---|---|
| Datasource | Selects which monitoring data source the dashboard queries. Switch it when you run more than one monitoring backend or environment and need to inspect one in isolation; every KPI, trend, and agent status on the page re-points to the chosen source. | single-select dropdown | No | LoadGen-Monitoring-Influx (name truncated in the UI) |
| Profiles | Narrows the dashboard to one or more End-to-End Monitoring profiles. Leave it at All to see every monitored flow at once; select specific profiles to isolate one flow's health while troubleshooting. | multi-select dropdown | No | All |
| Tests | Narrows the view to specific tests within the selected profiles. Use it after narrowing by Profiles to drill into a single scripted test instead of an entire flow. | multi-select dropdown | No | All |
| Measurement Filter | Limits which captured measurement actions feed the KPI cards and Trends chart. Selecting fewer measurements focuses the trend line on one step of the flow instead of an averaged view across all steps; the counter shows how many of the available measurements are currently selected. | multi-select dropdown, shows selected/total count | No | 0/0 (no measurements available yet in this environment) |
| Auto | Turns automatic refresh on or off. Checked, the dashboard re-queries on its own at the Refresh interval; unchecked, the view stays frozen until you click Refresh, which is useful when you want to read numbers without them changing mid-check. | checkbox | No | checked |
| Refresh interval | Sets how often the dashboard re-queries while Auto is on. A shorter interval gives a closer-to-live view at the cost of more frequent backend queries; a longer interval suits a dashboard left open on a wall display. | dropdown | No | 30s |
Column reference
| Column | What it shows | How to read it |
|---|---|---|
| Agent | Identifies which configured LoadGen Agent the row reports on. | Matches the agent name configured on Configuration: End-to-End Monitoring: Platform Agents; use it to find the right agent to enable or investigate. |
| Status | Shows the agent's current reporting state. | Online means the agent is active and reporting; Delayed means it reported later than expected; No Data means it is registered but has not reported anything yet; Disabled means it has been turned off and will not run sessions. |
| Last Run | Shows the timestamp of the agent's most recent completed run. | A dash means the agent has not completed a run yet in the selected time range. |
Controls reference
Time range selector (1h / 4h / 8h / 24h / 7d / Custom)
Sets the look-back window used by the KPI cards, the Trends chart, and the session list.
When to use: Use 1h or 4h to watch a run as it happens, 24h or 7d to spot drift or recurring failures over time, and Custom to pick an exact start and end.
Side effects: Read-only: re-queries the page for the new window, no data is changed.
Refresh
Re-queries the dashboard immediately with the current filters and time range.
Side effects: Read-only: updates the KPI cards, Trends chart, and the Updated ... ago timestamp.
Trend metric selector (Mean / P95 / Success %)
Switches which statistic the Trends chart plots.
When to use: Use Mean for typical response time, P95 to see tail latency affecting your slowest sessions, and Success % to track the pass or fail rate over time instead of timing.
Side effects: Read-only: redraws the Trends chart only.
Dismiss credentials banner
Closes the Credentials expiring banner.
When to use: Use it once you have noted the warning or already reissued the credential.
Side effects: Hides the banner for the current view; it does not reissue or fix the underlying credential.
Profiles dropdown
Opens a multi-select panel for the Profiles filter; a Show All link resets the selection to every profile.
When to use: Use it to choose which monitored profiles' sessions appear on the page.
Side effects: Read-only: changes which data populates the KPI cards, Trends chart, and session list.
Tests dropdown
Opens a multi-select panel for the Tests filter, scoped to the tests belonging to the selected profiles.
When to use: Use it after narrowing by Profiles to isolate a single scripted test.
Side effects: Read-only: changes which data populates the page.
Measurement Filter dropdown
Opens a multi-select panel listing captured measurement actions, with Select All and Deselect All shortcuts.
When to use: Use it to focus the Trends chart on one or a few measurement steps instead of the full set.
Side effects: Read-only: changes which measurements feed the Trends chart and KPI cards.
Session AI analysis toggle
Opens or closes an inline AI Analysis panel for an individual session row in the session list.
When to use: Use it once sessions are listed, to get an AI-generated read on why a specific session failed or behaved unexpectedly.
Side effects: Opens a panel for that row; closing it with its own close control does not change the underlying session data.
Sessions pagination (Previous / Next)
Moves between pages of the session list below the Trends and Agents panels.
When to use: Use it to browse past sessions beyond the first page.
Side effects: Read-only: changes which page of session rows is displayed.
Verify the result
On a healthy environment, the Sessions, Success Rate, Errors, and Measurements KPI cards show non-zero activity for the selected time range, Data Freshness shows a recent timestamp instead of No data, and the LoadGen Agents panel shows at least one agent as Online. The Trends chart plots a line for the selected metric (Mean, P95, or Success %) instead of the No trend data available message. If every KPI card still reads zero after widening the time range to 24h or 7d, check first whether any LoadGen Agent is Online and whether a profile is actually scheduled to run.
Troubleshooting
- Symptom: All KPI cards show 0 and Data Freshness reads No data. Fix: Confirm a profile is running on an enabled LoadGen Agent, or widen the time range to 24h or 7d before assuming nothing ran.
- Symptom: An agent in the LoadGen Agents panel shows Disabled with no Last Run. Fix: Enable the agent on Configuration: End-to-End Monitoring: Platform Agents; a disabled agent cannot execute monitored sessions.
- Symptom: Selecting a Profile or Test does not change the numbers on the page. Fix: Reopen the dropdown to confirm the selection took effect, then check that the chosen profile actually has sessions in the active time range.
- Symptom: The Trends chart still shows No trend data available even though the KPI cards show activity. Fix: Check the Measurement Filter count; it may be narrowed to 0 selected measurements, so widen or clear it.
- Symptom: The Credentials expiring banner keeps appearing after reissuing the token. Fix: Dismiss it with the X, then refresh the dashboard; reissuing on Configuration: Credential Profiles does not retroactively clear an already-rendered banner.
- Symptom: KPI numbers and the Updated ... ago timestamp stop changing. Fix: Confirm Auto is checked with a Refresh interval set, or click Refresh manually.
Related articles
- Prerequisite: End-to-End Monitoring: Profiles - profiles must exist and run before sessions appear on this dashboard
- Prerequisite: Configuration: End-to-End Monitoring: Platform Agents - agents must be configured and enabled to run and report sessions
- Prerequisite: Configuration: Datasources - the dashboard queries data from a connected monitoring datasource
- See also: End-to-End Monitoring: Schedules - schedules determine when monitored sessions run and feed this dashboard
- See also: Configuration: Credential Profiles - reissue credentials flagged as expiring in the banner
- Next step: Service Levels: Dashboard - ongoing monitoring health here feeds broader service-level tracking