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
Thank you for your answer. Before opening the issue, I had gone back and forth between concurrent-ruby 1.1.6 and 1.17 several times, and I was able to reproduce the error.
I took several steps to investigate: I used Rails 6.0.3.2 instead of 6 alpha; I forked concurrent-ruby and reverted #879; I changed several assets to force new compilation; and under all these circumstances I could not duplicate the problem. So it is difficult to say what the cause. So I shall have to apologize for the trouble.
In the end, I am back where I started, namely 1.1.7 and Rails 6 alpha. Since I could not duplicate my own problem, perhaps there was something unexpected about the particular assets that were being compiled. So changing to 1.1.7 pointed out a problem that I didn’t know I had that I was able to fix, but I still don’t know what it was.
Encountered two errors using webpacker and Rails 6 alpha when compiling assets onto a Centos 7 system after upgrading concurrent-ruby to v 1.1.7.
OS: Linux Centos 7
Ruby: ruby 2.7.1p83 (2020-03-31 revision a0c7c23c9c) [x86_64-linux]
Rails: 6-alpha
concurrent-ruby-ex and concurrent-ruby-edge not installed.
The first problem was NoMethodError: undefined method `+' for nil:NilClass. This error was solved (see below).
This was solved by adding an additional paths declaration to config/webpacker.yml
additional_paths: []
The next deploy, a segmentation fault error appeared:
The full backtrace is available if relevant. Reverting to 1.1.6 solved this issue.
We have other projects using Rails 6.0.3.2 and the asset pipeline that are not affected by upgrading to 1.1.7.
The text was updated successfully, but these errors were encountered: