Skip to content
Opentra
Changelog

What's new in Opentra.

You depend on this software to run your MSP. So every release gets real notes — what shipped, what changed, what got fixed.

13 August 2026

v2026.08.11
Fixed

Start a ticket from a contact and the details are already filled in

When you click New ticket from a contact or org page, the ticket form now arrives with the Contact and Company already set, so you no longer have to re-search for the person you were just looking at. Archived contacts or orgs are filtered out, so you won't get a blank field that still submits. If a contact belongs to a different org than the one in the link, the contact's own organisation wins, so the two can never disagree.

NEW TICKET (from contact)
Contact: Sarah Whitlam
Company: Northbridge Dental
✓ pre-filled, no re-search

7 August 2026

v2026.08.8
Changed

Grab planner cards from anywhere

You can now drag a planner card by grabbing it anywhere, not just the tiny six-dot grip. The whole card is the drag handle, so moving work around the board is quicker and the affordance is obvious. A plain click still opens the ticket, project, or manual card edit dialog, because a drag only starts once you actually move the card, so navigation and dragging no longer fight each other.

PLANNER CARD
#512 VPN drops
grab anywhere to drag
Improvements and fixes (1)
  • The done checkbox and Remove button no longer trigger an accidental card drag when you tap them

4 August 2026

v2026.08.6
Fixed

Client email attachments now stick to org and opportunity timelines

When a client replied to a CRM or org-level email with a file attached, the reply landed on the activity timeline but the attachment quietly went missing. That path never saved the file, even though it was pulled down correctly. Now those inbound replies keep their attachments on both opportunity and org-level activities, stored the same way ticket replies always have been. Signature logos and inline images are filtered out so your timeline shows real files, not logo chips.

ORG TIMELINE · INBOUND REPLY
Josh Saliba: "See attached"
forecast.xlsx
✓ attachment kept on timeline
Added

Recover attachments that went missing before the fix

Replies received before this fix never got their attachment records, but the files still live in your mailbox. A re-runnable backfill script finds those inbound CRM and org activities, re-locates each message in Microsoft 365, and restores the missing attachments to the timeline. It is idempotent, supports a single-activity run, and has a dry-run mode so you can see what it will touch before it touches anything.

Improvements and fixes (1)
  • Activity attachments now serve through the same auth-gated proxy as ticket attachments, so they open cleanly in the browser instead of pointing at an un-servable link

3 August 2026

v2026.08.148 – .3
Fixed

Monthly reports now count only the devices you actually monitor

Coverage gauges used to assume 100% of every device, counting machines that had never checked in, plus hand-entered inventory with no agent at all. Now monitoring and security coverage are measured against agent-backed devices only, using real check-in data (last seen within 30 days). Hand-catalogued kit no longer reads a red "0 of 11", and a client with 4 dark devices shows a truthful figure like 15 of 16 instead of a false all-clear. The narrative and portal donut both use the honest number.

MONITORING COVERAGE
19 of 19 (100%)
→
15 of 16 (79%)
✓ counts devices actually reporting
Fixed

Monthly reports no longer count monitoring alerts as support requests

Auto-generated RMM alert tickets were being lumped in with genuine support, so one client's report claimed "we raised 143 tickets" when only 35 were real requests and the other 108 were monitoring auto-tickets (73 from a single flapping sensor). Support counts, first-response averages and contacts helped now count email and manual tickets only. Monitoring volume is reported separately, and de-duplicated so one noisy sensor collapses to a single distinct issue (108 raw down to 24) instead of looking like many problems.

JULY ACTIVITY
Support requests: 35
Monitoring alerts: 24 distinct
was: one report of "143 tickets raised"
Fixed

Fixed-fee clients no longer get dropped from a billing run

Billing runs used to be built purely from activity in the period: approved billable time, product lines, projects and milestones. A client whose only charge was a fixed monthly fee (an MSA base fee, a retainer, or block hours) got silently left out unless they happened to have other activity that month. Now the run pulls in every org with an active recurring or fixed-fee contract, so a 19-contract book stops producing a 12-client run. There is no double-billing risk: lines are still computed fresh at review and finalise from live contract data.

Improvements and fixes (10)
  • Monthly reports skip months that predate the client, so a mid-month signup no longer gets a glowing report for a period it wasn’t yet a client
  • Empty orgs (no tickets, time, alerts, devices or renewals) no longer send a hollow “we did nothing” report
  • Manual “Send test / Send now” now targets the client’s latest reportable month and gives a specific skip reason instead of a vague catch-all
  • “Resolved” support counts now include tickets sent straight to Closed, ending phantom backlogs and invented carry-over language
  • EDR coverage now reflects agents actually installed and reporting, not just devices that had previously raised an alert
  • Huntress agent to device matching now tolerates FQDN and .local hostnames, so agents stop failing to link to the right machine
  • Coverage figures now flag network gear and printers that can’t run an agent, so the report stops implying they should
  • Level.io sync now adopts a matching manually-added device instead of leaving two cards for one physical machine
  • Fixed stacked duplicate “Show quoted history” buttons piling up inside email ticket bodies
  • Fixed a project reply box that didn’t clear after sending, which was causing duplicate replies

29 July 2026

v2026.07.146 – .147
Added

Reschedule planner work you didn't get to

Work you planned but never got round to no longer disappears. Anything left unfinished from an earlier day now surfaces in an Overdue strip at the top of this week's planner. Drag a card straight onto a new day, or hit "Roll all to today" to sweep the whole backlog forward in one click. You can also drag a card onto the week arrows to bump it to the same day next week (or last week), so rescheduling across weeks is a single drag.

OVERDUE
Swap failed drive at Acme, was due Mon
Roll all to today
then drag to any day
Fixed

Recovery alerts now close the ticket they belong to

When a monitored device recovers, the "Resolved" alert email used to open a brand new ticket instead of closing the one it belonged to, leaving you with a stray open ticket for a problem that had already cleared. Recovery notices are now matched to the original incident and close it out. A repeat alert for a still-open incident is merged onto the existing ticket, and if it comes in hotter than before, the priority and response deadline are raised to match.

ALERT LIFECYCLE
Critical: low disk
->
Resolved
closes ticket #838, no duplicate opened
Improvements and fixes (2)
  • Alert tickets now inherit the alert’s real severity as their priority, so a critical incident lands as critical from the start instead of a flat medium
  • A recovery notice with no incident on record is quietly dropped instead of opening a ticket for a problem that is already over