Monitoring dashboard
The Monitoring dashboard gives site admins a live view of how Foundation is running inside their Jira tenant. Use it to spot slow resolvers, rate-limit pressure, and recurring errors before they turn into user complaints.
You need Jira Administer Jira permission to open the dashboard.
Open the dashboard
Section titled “Open the dashboard”- From Apps > Manage your apps > Foundation, click Configure.
- Go to the Monitoring tab.
- Pick a time range: Last hour, Last 6 hours, or Last 24 hours.
The page refreshes on open and stays current as long as it’s visible.
Resolver performance table
Section titled “Resolver performance table”The main table shows one row per backend resolver (API endpoint) with:
- Name — resolver identifier (for example,
getLensView,updateField). - Count — total invocations in the selected window.
- Avg ms — average execution time.
- p95 ms — 95th-percentile latency — a better signal than average when a few slow calls matter.
- Error count — number of failed invocations.
- Error rate — percentage of calls that errored.
Click any column header to sort. Sort by p95 ms or Error rate when hunting for trouble.
Jira API point history
Section titled “Jira API point history”A bar chart shows the Jira API points Foundation consumed each hour against your available budget. Foundation is on Tier 2 (per-tenant) so most sites have plenty of headroom, but spikes — often a large Sync Agent run — are visible here before they cause rate-limit errors.
Recent errors
Section titled “Recent errors”The bottom of the page lists the most recent backend errors: resolver name, error message, and timestamp. Use this to correlate user reports with backend failures.
Pruning
Section titled “Pruning”Monitoring data accumulates in Forge KVS. Click Prune to drop records older than seven days. This is automatic on a schedule, but manual pruning is available if you need to shrink storage quickly.
What to watch for
Section titled “What to watch for”- Consistent p95 ms above a few seconds on
getLensView— probably a lens with too many issues, or a cold issue cache. - Error rate above ~1% on
updateField— usually a workflow validator is rejecting a common bulk-edit pattern. - API points trending toward the budget — consider splitting large JQL agents into smaller scopes.