v0.38.0: an ITIL helpdesk
Until now the helpdesk was a solid ticket system: an email inbox, ticket lists per customer, SLA deadlines, macros, and checklists. v0.38.0 turns it into an ITIL service desk. It’s the biggest release so far, so here’s the tour.
Incidents
Tickets get two new statuses. In progress means someone is working on it; Resolved means the fix has been delivered. Resolved tickets close automatically after a few days (five by default) unless the requester replies, and a reply reopens the ticket and counts as a reopen.
Instead of picking a priority, set Impact and Urgency: the priority follows from a matrix an administrator can edit, and you can still override it. Tickets also get two-level categories, and resolving one asks for a resolution code and notes, which an administrator can make mandatory. Flag a ticket as a Major incident and a chosen group is alerted straight away.
Ticket links now have a type (related, duplicate of, caused by, parent of), and you can link to another customer’s ticket as long as you have access to it.
SLAs that count working hours
SLA deadlines can now be counted in business hours: give a policy a calendar with working days, hours, a time zone, and holidays (one click imports a country’s public holidays). While a ticket waits for the requester, the resolution clock pauses. A policy can apply to a customer, a ticket type, or a category, and the most specific one wins.
The server now checks every open ticket once a minute. When most of the SLA time is used, the assignee and group get a warning; on a breach the ticket is escalated to the policy’s escalation group or user, and reassigned to them unless you switch that off.
This release also fixes a few SLA bugs: tickets closed late are now always marked as breached, raising the priority no longer hands out a fresh deadline, and tickets from email get an SLA too.
Problems and the known error database
When several incidents share a cause, open a problem. Record the root cause and a workaround, mark it as a known error, and link the incidents to it with caused by. Once it’s fixed, resolve all linked incidents at once with a message to each requester. The new Problems page lists every problem and doubles as the known error database, and an agent can paste a problem’s workaround into a reply with one click.
Knowledge base
Write how-to articles and solutions in Markdown. Internal articles are for agents; public ones also appear in the customer portal, for every customer or just one. Every ticket shows matching articles, and Create article from this ticket turns a resolution into an article in seconds.
Service catalogue and approvals
Administrators can now publish orderable services, such as a new laptop or access to a project, each with its own request form, fulfilment group, and task checklist. Ordering one creates a service request with the answers, routed to the right group, and customer-portal users can order too.
Requests that cost money can need approval from a group: set how many members must approve, and a single rejection rejects. Until then nobody can start on it. Approvers get an Approvals page with all the details they need.
Changes
Change requests now follow a proper workflow: draft, awaiting approval, approved, scheduled, implementing, review, and completed or failed. A change has a type (standard, normal, or emergency), a risk level, a planned window, and implementation, backout, and test plans. Normal and emergency changes need approval from the change advisory board; standard changes are pre-approved. Freeze windows block scheduling (except for emergencies), and WarmDesk warns when another change touches the same configuration items in the same window.
The new Change calendar shows it all in a month view, a week view with a time grid, or a list.
CMDB
Finally, a configuration management database: keep track of the services, applications, servers, databases, and network gear you support, per customer or internal. Relations such as depends on and runs on drive an impact analysis that shows everything affected when an item fails, with its open tickets. Link affected items to tickets, and a change shows the combined impact of its items.
Ticket report
A new Ticket report page shows the numbers: tickets created and resolved, backlog and its age, mean time to resolve and to first response, SLA compliance, first-contact resolution, and reopen rate, broken down by priority, type, category, and customer. Export it as PDF or Excel.
warmdesk-seed creates demo data for all of the above, so warmdesk-seed --reset is the quickest way to try it.
See the Helpdesk sections of the User Guide and the Admin Guide for the details.
Upgrade
Download v0.38.0 from the download page or pull the latest release tag. The new tables and columns are created automatically on startup; tickets that were already resolved or closed get their resolution times filled in from their history on the first start.