Hendrik Liebau

@kaizencc Is this still on your radar? Also, I can't find any info that Node 20 is only available in certain regions (e.g. https://docs.aws.amazon.com/lambda/latest/dg/lambda-runtimes.html, or https://aws.amazon.com/blogs/compute/node-js-20-x-runtime-now-available-in-aws-lambda/). Can you confirm that...

Maybe we should use `Promise.allSettled` here: https://github.com/feature-hub/feature-hub/blob/11490453001ea2628ecb9aa291d494618f429305/packages/async-ssr-manager/src/internal/async-ssr-manager.ts#L72

* **#80463** 👈 (View in Graphite) * **#80462** * **#79743** * **#79629** * **#80461** * `canary` This stack of pull requests is managed by Graphite. Learn more about stacking.

@tomwanzek I think this is a misunderstanding. Jimmy didn't specifically answer this question by Colin: >Is it based on the URL a server action is invoked from? The answer to...

@MarkLyck This should be fixed in Next.js 15.0.0-canary.111. Let me know in case you still notice issues with stream buffering.

Putting this back on hold, for some reason the fix does not work anymore. Might revisit later.

The observation that the fix suddenly didn't work anymore was grounded in the fact that I originally tested with an app that was using the edge runtime (which worked), and...

* **#80463** * **#80462** 👈 (View in Graphite) * **#79743** * **#79629** * **#80461** * `canary` This stack of pull requests is managed by Graphite. Learn more about stacking.

Totally forgot to mention: This would also explain why increasing the `maxLength` did fix the problem as well. Because then there are no ID collisions, and the deterministic ID assignment...

Converting it to a draft while I investigate the two test failures.