Degraded service affecting 4 components

Update timeline

1. investigating Aug 27, 2026, 04:56 AM UTC

   We are investigating an issue affecting Vendors service, Vendor portal, Website & marketplace and Catering ordering & quotes. You may see errors or missing information on these parts of Feedr. We are working on it and will update this page as we learn more.

2. resolved Aug 27, 2026, 04:59 AM UTC

   All affected services have returned to normal. A full incident report follows.

3. postmortem Aug 27, 2026, 04:59 AM UTC

   **Incident report: reduced performance, 27 August 2026** Between approximately 05:49 and 05:59 BST, parts of Feedr were slower than usual and some pages did not load correctly. We are sorry. Ordering lunch should be one of the least eventful things in your day, and for a few minutes it was not. **What happened** Early that morning, demand on the systems that serve supplier and menu information rose sharply, to roughly double what we would expect at that time of day. While those systems scaled to meet the increase, requests took longer to process than normal, and some did not complete within the time our services allow before giving up. That is why some pages were slow, showed incomplete information or returned an error instead of loading. Demand returned to normal within a few minutes and performance recovered on its own, without any intervention needed. Nothing was lost, altered or left in a broken state at any point. Two characteristics of our current setup meant the spike was felt more than it should have been. The database serving this information does not yet have a second copy to share read traffic across, and one of our services opens a new database connection for each request rather than drawing from a shared pool. Both are scaling characteristics rather than faults, and both are now under review. **What this meant for you** For the duration, and depending on which part of the product you were using, you may have seen slower page loads, incomplete supplier or menu information or an error where a page would normally appear: * Website & marketplace * Catering ordering & quotes * Supplier information * Vendor portal A small number of individual requests did not complete and would have shown an error at the time. Trying again would have worked normally once performance recovered. Sign-in, checkout and card payments were unaffected and remained available throughout. No data was lost, altered or exposed. **Timeline** | Time | Event | | --- | --- | | 05:45 BST | Demand on the affected systems began rising above normal levels | | 05:49 BST | Some requests began taking longer than our services allow and returned errors | | 05:54 BST | Our monitoring detected the problem and opened this incident | | 05:57 BST | We updated this page to confirm we were investigating | | 05:59 BST | Demand had returned to normal, performance had recovered and the incident was closed | **What we are doing about it** Everything recovered on its own within a few minutes, but we would rather it had not reached you at all. * We are adding monitoring of database load directly, covering both connection volume and processing capacity, so we can see demand building before it affects anyone. In this case the pressure was visible several minutes before any page was affected. * We are investigating our scaling requirements for these systems as demand grows, including adding a second database copy to share read traffic and pooling connections for the service that opens them individually. * We are widening what our monitoring covers, so that more of the ordering journey and more of the regions we operate in are checked continuously rather than inferred from the services behind them. We will continue to monitor demand patterns and review capacity ahead of growth rather than in response to it. **If this affected you** If your organisation missed an ordering deadline or you have any concern about an order placed around this time, please contact your Feedr account manager or our support team and we will investigate individually. Thank you for your patience. We would rather tell you plainly what went wrong than not mention it at all.