Jira Cloud

1. Sign in to Atlassian and go to https://id.atlassian.com/manage-profile/security/api-tokens.
2. Click Create API token. Give it a label that identifies this integration (e.g. Ingest connector).
3. Pick an expiration window. Atlassian forces a 1-year maximum; tokens must be rotated before expiry or the connector will return 401. Note the expiration date and set a reminder.
4. Copy the token. Atlassian only shows it once — if you close the dialog without copying, you'll need to recreate it.

1. In the Ingest UI under Connectors → Jira Cloud, paste the bare subdomain into Atlassian Cloud subdomain (e.g. acme for https://acme.atlassian.net).
2. Paste your Atlassian account email into Email.
3. Paste the API token into API token.
4. Save. The connector validates by calling /rest/api/2/myself against your tenant; a successful response confirms credentials and tenant subdomain.

API-token traffic is governed by Atlassian's burst rate limits only — the broader points-based hourly quota explicitly excludes Basic auth callers. Burst defaults to 100 requests/second per endpoint per tenant, with select hot endpoints (e.g. /issue/{id}/changelog) raised to 200 req/s. The connector throttles globally to 5 req/s, well below the floor.

The runtime treats 429, 500, 502, 503, and 504 as retryable and respects the Retry-After header. 400, 401, 403, and 404 are fatal — the most common cause of 400 is an unbounded JQL query (the new /search/jql endpoint requires at least one criterion) or a mismatched endpoint version. 401 means the token is wrong, expired, or the email doesn't match the token owner.

The connector ships 39 platform endpoints plus 2 admin/paid endpoints that stay disabled by default. Most analytics use cases need this core set:

• Issues — issue_search is the workhorse (cursor-paginated via /rest/api/2/search/jql); pair with issue_changelog, issue_comments, and issue_worklog for full ticket history.
• Projects — project_search plus project_versions and project_components for release planning and component breakdowns. project_categories and project_types are tiny dimension tables.
• Issue dimensions — issue_types, statuses, status_categories, priorities, resolutions, issue_link_types, and fields give the lookup tables you need to translate raw issue data into reports.
• Boards & sprints (Jira Software customers only) — boards, board_sprints, board_issues, sprint_issues, and board_configuration. Free tier of Jira Software is enough.
• Workflows & schemes — workflows, workflow_schemes, permission_schemes, notification_schemes, plus four scheme-related lookups. Most customers enable a subset based on their reporting needs.
• Users & groups — users_search, groups, group_members.
• Dashboards & filters — dashboards, filters, plus filter_my and filter_favourite for per-user views.
• Sysinfo — myself, server_info, my_permissions for identity and permission auditing.

audit_records (admin scope only) and priority_schemes (Premium/Enterprise edition only) ship under untested/ and need manual activation per your account's plan and permissions.