You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Looking at pipelines triggered by merging module updates in the bcr, I noticed that the on-bcr-trigger pipeline regularly runs into timeouts, leading to updates not being available until the next commit.
Error message:
> Build error occurred
Error: Static page generation for /modules/abseil-cpp is still timing out after 3 attempts. See more info here https://nextjs.org/docs/messages/static-page-generation-timeout
A simple fix could be to increase the timeout and see if it helps.
The text was updated successfully, but these errors were encountered:
Without having profiled it, my best guess for the failure is:
The map that contains all module versions (required on all pages due to the dependencies section) and their metadata takes too long to build. It's cached across page builds, but the build via a shared promise for all pages, which is why they will all timeout once assembling that map crosses the threshold of 60s
The slowest operation for each module version is determining the release date/the BCR commit that introduced that version. That's quite slow, as it has to traverse the git history
To prevent hitting the time limit of 1., the assembly for that metadata could be pulled into a seperate build step that dumps the data in .json file at a known location, which is then picked up by the Next.js build (this should also have the side-effect of speeding up hot page reloads during development of the UI).
To speed up 2. we could also try to cache that slow to calculate data between builds, especially as it shouldn't change over time, once a release is added to the BCR.
Looking at pipelines triggered by merging module updates in the bcr, I noticed that the
on-bcr-trigger
pipeline regularly runs into timeouts, leading to updates not being available until the next commit.Error message:
A simple fix could be to increase the timeout and see if it helps.
The text was updated successfully, but these errors were encountered: