Homepage fails for Googlebot: stale JS chunk no longer exists after deployment

Error

ChunkLoadError: Failed to load chunk /_next/static/chunks/2h5_roi10jupk.js from module 964893

Stack trace

entry (turbopack/browser/runtime/base/runtime-base.ts:233)

Source (turbopack/browser/runtime/base/runtime-base.ts:228-238)

          invariant(
            sourceType,
            (sourceType) => `Unknown source type: ${sourceType}`
          )
      }
      let error = new Error(
        `Failed to load chunk ${chunkUrl} ${loadReason}${
          cause ? `: ${cause}` : ''
        }`,
        cause ? { cause } : undefined
      )

Request context

• URL: https://main.devguard.org/
• Culprit: entry(chunks/turbopack-26bafpkzoigym)
• User agent: Googlebot 2.1 crawler
• Classic stale-build issue: the crawler loaded an HTML page referencing chunk 2h5_roi10jupk.js, but by the time it requested that chunk the build had moved on (new deployment replaced static assets), so the file no longer exists at that path. Not an app logic bug — a side effect of rolling deploys invalidating old chunk hashes while old page references are still being crawled/cached.

GlitchTip

• Issue: https://error-tracking.devguard.org/devguard/issues/3831
• Event ID: 019fc52bbf2d7426a351f4135aa74612
• First seen: 2026-08-03T00:00:00.040Z

Start at turbopack/browser/runtime/base/runtime-base.ts:228-238 and inspect how rolling deployments serve HTML and static chunks. Use the reported homepage, stale chunk name, and Googlebot context to trace the failure between page load and chunk retrieval. Done means delayed or cached page requests no longer produce this missing-chunk failure, with a regression check if the repository provides one.