Closed beku8 closed 6 years ago
We have created an issue in Pivotal Tracker to manage this:
https://www.pivotaltracker.com/story/show/156364507
The labels on this github issue will be updated when the story is started.
Thanks for the report! We have prioritized an investigation.
@beku8 can we get some information about what you are doing in the postinstall? An example script would be helpful.
@beku8 This change you're seeing in the directory from which npm install
(and hence postinstall
) runs is intentional and expected (in order to avoid modifying the app directory during supply). That said, we have recently decided to go another route with this particular buildpack. As a consequence, you should see your issue resolved in the next buildpack release.
@ameyer-pivotal good to hear that!
@tylerphelan our script is "postinstall": "NODE_ENV=production next build"
. This enables us to upload a small package and build it on cloudfoundry runtime.
Fixed in v1.6.22.
Yes its working properly now, thanks!
We have a nodejs (nextjs to be precise) app that builds itself using the
postinstall
script which gets triggered after thenpm install
happens.After the
1.6.21
release our build was failing on cloudfoundry, so I did a diff on the console output and found out that it was working from a different directory than the previous releaseon 1.6.20,
> ui2@1.0.0 postinstall /tmp/app
on 1.6.21> ui2@1.0.0 postinstall /tmp/contents058201511/deps/0/packages
Is this intended feature or can it be fixed back to the original?