A separate review confirmed that the storefront displayed correct savings estimates on retest. This was a calculation validation milestone, not an extension of the availability disruption resolved at 16:24 UTC−3.

The savings discrepancy involved incorrect utility and tariff information. The storefront workaround retrieved location-specific utility data separately. Correction of utility data persisted in Artemis remained a separate follow-up item.

Successful project creation and design display were confirmed in the affected storefront.

Project records submitted during the disruption were retained in Artemis. Affected users needed to restart the quote creation flow to view their designs.

This timestamp marks the verified recovery of the storefront availability issue.

The design service endpoint configuration was corrected. The team then tested new quotes to verify that project creation and design display worked through the storefront.

The investigation identified an outdated design service endpoint returning HTTP 503 errors. The storefront integration still referenced a development endpoint that had been retired during a service migration.

This prevented the storefront from retrieving and displaying designs.

A recently changed storefront domain was added to the authorized domain configuration. This addressed a configuration issue, but subsequent tests showed that the quote creation and design display failures persisted. Investigation continued.

This retrospective report documents a storefront disruption on October 9, 2026. Affected users encountered errors or blank pages when creating quotes and viewing designs.

The disruption began at approximately 07:30 UTC−3. The endpoint correction was applied at 16:13, and recovery was verified at 16:24. A separate savings calculation discrepancy was also reviewed that day.

The team received reports that the affected storefront displayed a blank page or an error during quote creation. Projects were being recorded in Artemis, but users could not view their designs.

The subsequent incident review placed the start of the disruption at approximately 07:30 UTC−3.