Sage 50 and Jira integration

What moves between them

Data flows primarily from Jira into Sage 50. Issue records are read from Jira via REST API, matched to Sage 50 projects by custom field mapping, and optionally create or link general journal entries and project expenses in Sage 50. Worklog entries from Jira are polled and can populate Sage 50 project timesheets or employee hours. Project metadata (codes, names, status) can flow from Sage 50 to Jira via custom fields to keep project visibility consistent across both systems.

How ml-connector handles it

ml-connector runs on a Windows machine with Sage 50 installed and SDK access available. It accepts Windows local credentials (username, password, ApplicationID or DataPath per edition) encrypted and stored per cell, authenticates to Jira via OAuth2 bearer token, and polls Jira for issues and worklogs on a configurable schedule (typically every 15 minutes). Issues are matched to Sage 50 projects via a custom field in Jira that stores the Sage 50 project code. For each issue, ml-connector validates that the project code exists in Sage 50 before creating a linked general journal record or project expense line. Worklog entries are read from Jira and mapped to Sage 50 project timesheets or employee records. Jira webhooks, when enabled, can push issue and worklog events to ml-connector; however, because webhooks expire after 30 days, ml-connector also polls Jira to catch any missed events. The local Sage 50 SDK process requires an interactive Windows session and exclusive user access, so ml-connector runs as a scheduled task that polls incrementally and exits, avoiding long-running sessions that would lock the SDK. Every record carries an audit trail and can be replayed if a downstream write fails.

A real-world example

A mid-sized services firm uses Sage 50 on a local desktop for invoicing, expenses, and project accounting, and Jira Cloud for task and issue management across project teams. Before integration, issues in Jira were created by project managers, but finance had no visibility into which Sage 50 projects were being worked on and had to manually reconcile timesheets from outside Jira. With Sage 50 and Jira connected, each issue in Jira includes a project code custom field. ml-connector reads those issues, validates the project exists in Sage 50, and creates a project expense record with the linked issue ID. When team members log work in Jira, ml-connector captures the worklog and can optionally post it to Sage 50 project timesheets. Finance now sees real-time issue work mapped to Sage 50 projects, eliminating manual timesheet entry and keeping project accounting accurate.

What you can do

• •Link Jira issues to Sage 50 projects using a custom project-code field, validating the project exists before creating records.
• •Capture worklog entries from Jira and sync them to Sage 50 project timesheets or employee hour records.
• •Poll Jira on a configurable schedule and handle webhook refresh cycles to prevent event loss.
• •Authenticate to Sage 50 via local Windows SDK (supporting both US .NET and UK COM editions) and Jira via OAuth2 bearer token.
• •Create and maintain a full audit trail of all linked issues, worklogs, and project expense records, with replay capability on failure.

Questions

How does ml-connector authenticate to both Sage 50 and Jira?

Sage 50 authentication is Windows-local: ml-connector stores the Windows username and password encrypted and uses them via the SDK (ApplicationID and CompanyPath for US edition, or DataPath for UK edition). Jira authentication is OAuth 2.0: ml-connector exchanges an authorization code for a bearer token and uses it for all REST calls. Both credential sets are encrypted per cell.

What happens if Jira issues reference Sage 50 projects that do not exist?

ml-connector validates every project code in Jira against the list of active projects in Sage 50 before creating any linked record. If a project does not exist, the record is logged as a failed record with the reason (project not found) and can be replayed once the project is created in Sage 50.

Why does ml-connector poll Jira if Jira webhooks are available?

Jira webhooks expire after 30 days and must be manually refreshed. ml-connector enables webhooks where available but also polls Jira on a schedule to catch any events that may have been missed during a webhook expiry window, ensuring no issue or worklog updates are lost.