A Stability Follow-Up
Yesterday's release stopped the duplicate refresh work behind the Error 1102 failures on Quotes and Invoices. This release takes the next step: it removes the remaining automatic timer that refreshed list pages even when nothing had changed.
Quotes and Invoices will still catch up when you return to a tab, so switching back to Current brings the page up to date. What they no longer do is repeatedly ask the server for the same data simply because a clock ran out. That means less invisible work from every open tab and more room for the actions you are actually taking.
Better Protection Against Runaway Requests
We also tightened the safety limit around individual server requests. Normal pages use only a fraction of the available processing time, but the previous configuration allowed one runaway request to keep consuming resources for far too long.
The new limit leaves substantial headroom for sign-in, page rendering, and larger records while stopping abnormal work much sooner. This is a guardrail rather than a speed limit: normal requests should behave exactly as before, while a broken or unexpectedly expensive request is contained before it can have a wider impact.
Security Maintenance
The application's production dependencies have been refreshed, including authentication, background jobs, and server-rendering components. This clears all known high and critical production dependency advisories present during the release audit.
The background-job deployment tools were upgraded in lockstep as well, so the task code running in production stays aligned with the version shipped in the application.
What You Should Notice
There is no new workflow to learn in this release. The intended result is quieter:
- Quotes and Invoices perform less unnecessary work while left open.
- Returning to an existing tab still refreshes its data.
- Abnormal server requests are contained sooner.
- Production dependencies no longer carry known high- or critical-severity advisories from this audit.
We are continuing to watch request timing and resource usage so that this class of problem shows up in our monitoring before it becomes an error page for you.