Attempting node version fails

Briefly summarize the issues you have been experiencing.
“Attempting node version ‘v10.16.0’ from .nvmrc” hangs for several minutes, reaches a few percent or fails completely

Please provide a link to your live site hosted on Netlify

What have you tried as far as troubleshooting goes? Do you have an idea what is causing the problem?
The problem suddenly appeared without any changes to node packages or node versions in my setup.
I have tried adding a .nvmrc file to specify desired node version, and I have tried both the previous successful version and newer versions supposed to be available. All with the same result.

Do you have any other information that is relevant, such as links to docs, libraries, or other resources?

I’m having the same issue. It fails at downloading v10.16.0 every time :frowning:

I’m experiencing the same issue. Have tried several different node versions as well.

Same here. Any advice on fixing?

2:09:36 PM: curl: (56) GnuTLS recv error (-54): Error in the pull function.

2:09:36 PM: Binary download from https://nodejs.org/dist/v10.16.0/node-v10.16.0.tar.xz failed, trying source.

2:09:36 PM: grep: /opt/buildhome/.nvm/.cache/src/node-v10.16.0/node-v10.16.0.tar.xz

2:09:36 PM: : No such file or directory

2:09:36 PM: failed during stage 'building site': Build script returned non-zero exit code: 1

2:09:36 PM: Provided file to checksum does not exist.

2:09:36 PM: Failed to install node version '10'

2:09:36 PM: Error running command: Build script returned non-zero exit code: 1

2:09:36 PM: Shutting down logging, 2 messages pending

Same here. No idea how to fix.

Same here this is what I’m seeing across ALL site redeploys.

Version ‘10’ not found - try nvm ls-remote to browse available versions.
Version ‘8’ not found - try nvm ls-remote to browse available versions.

We have the same problem on our frontend.

2:14:23 PM: Version '8' not found - try `nvm ls-remote` to browse available versions.
2:14:23 PM: Failed to install node version '8'
2:14:23 PM: Error running command: Build script returned non-zero exit code: 1
2:14:23 PM: failed during stage 'building site': Build script returned non-zero exit code: 1
2:14:23 PM: Failing build: Failed to build site

I’ve had the same issue so I’ve asked support about this and I got following response:
“We’ve identified the issue as being related to nvm or nodejs and the incident their hosting provider is experiencing: https://www.cloudflarestatus.com/incidents/46z55mdhg0t5”. (edit: fixed link)

1 Like

Looks like it is fixed now. We had a successful deploy on our staging site. :fireworks:

The problem is gone for me too. Thanks for letting me know I wasn’t alone so I didn’t have to spend more time debugging this.

hi everyone, as mentioned above, this was an issue related to a third party and should be resolved now. Please let us know if you have any more trouble.

Hey @perry, is there an aftermath?

Our builds are failing with repeated npm errors of this kind for diff packages:

6:59:42 PM: Computing checksum with sha256sum
6:59:42 PM: Checksums matched!
6:59:45 PM: Now using node v10.16.0 (npm v6.9.0)
6:59:45 PM: Attempting ruby version 2.3.6, read from environment
6:59:46 PM: Using ruby version 2.3.6
6:59:46 PM: Using PHP version 5.6
6:59:46 PM: Started restoring cached node modules
6:59:46 PM: Finished restoring cached node modules
6:59:47 PM: Installing NPM modules using NPM version 6.9.0
7:00:00 PM: npm
7:00:00 PM:  WARN tar ENOENT: no such file or directory, open '/opt/build/repo/node_modules/.staging/webidl-conversions-f578d840/package.json'
7:00:00 PM: npm
7:00:00 PM: WARN tar ENOENT: no such file or directory, open '/opt/build/repo/node_modules/.staging/webidl-conversions-f578d840/README.md'
7:00:00 PM: npm WARN
7:00:00 PM: tar ENOENT: no such file or directory, open '/opt/build/repo/node_modules/.staging/webidl-conversions-f578d840/LICENSE.md'
7:00:00 PM: npm
7:00:00 PM: WARN tar
7:00:00 PM:  ENOENT: no such file or directory, open '/opt/build/repo/node_modules/.staging/react-dom-2b8261c0/umd/react-dom-unstable-fire.production.min.js'
7:00:00 PM: npm
7:00:00 PM:  WARN tar ENOENT: no such file or directory, open '/opt/build/repo/node_modules/.staging/bootstrap-3419c89a/dist/js/bootstrap.js'
7:00:01 PM: npm WARN tar
7:00:01 PM:  ENOENT: no such file or directory, open '/opt/build/repo/node_modules/.staging/zxcvbn-546799cc/src/main.coffee'
7:00:01 PM: npm
7:00:01 PM: WARN tar ENOENT: no such file or directory, open '/opt/build/repo/node_modules/.staging/zxcvbn-546799cc/src/matching.coffee'
7:00:01 PM: npm
7:00:01 PM: WARN tar ENOENT: no such file or directory, open '/opt/build/repo/node_modules/.staging/onfido-sdk-ui-70dedb4a/dist/reports/statistics.html'
7:00:01 PM: npm
7:00:01 PM: WARN tar ENOENT: no such file or directory, open '/opt/build/repo/node_modules/.staging/zxcvbn-546799cc/src/scoring.coffee'
7:00:01 PM: npm
7:00:01 PM:  WARN tar ENOENT: no such file or directory, open '/opt/build/repo/node_modules/.staging/zxcvbn-546799cc/src/time_estimates.coffee'
7:00:01 PM: npm
7:00:01 PM: WARN tar ENOENT: no such file or directory, open '/opt/build/repo/node_modules/.staging/zxcvbn-546799cc/.zuul.yml'
7:00:01 PM: npm WARN tar
7:00:01 PM:  ENOENT: no such file or directory, open '/opt/build/repo/node_modules/.staging/eslint-0536df2f/lib/rules/init-declarations.js'
7:00:01 PM: npm
7:00:05 PM: failed during stage 'building site': Build script returned non-zero exit code: 1
7:00:01 PM: WARN tar ENOENT: no such file or directory, open '/opt/build/repo/node_modules/.staging/eslint-0536df2f/lib/rules/jsx-quotes.js'
7:00:01 PM: npm

They were working earlier today.

Rebuilding with clean cache doesn’t help.

Can you provide a link to one of your failing deploys @stefanmaric? Thanks.

Shared them in private. :slight_smile:

Whenever we see this pattern in a build failure:

7:00:00 PM: npm
7:00:00 PM: WARN tar
7:00:00 PM:  ENOENT: no such file or directory, open '/opt/build/repo/node_modules/.staging/react-dom-2b8261c0/umd/react-dom-unstable-fire.production.min.js'

It is almost always one of two things:

  1. your package-lock.json needs to be rebuilt - it refers to some temporary files that are not present during our build. This is the usual reason for that type of error.
  2. Less common, you’ve for some reason committed your node_modules directory to git, which is an antipattern for building on Netlify. Let package-lock.json or yarn.lock do their jobs and specify versions for you and we’ll fetch them. Of course this won’t work well when Cloudflare is down, but that’s the main reason it should fail…

Indeed. :pensive:

It was actually related to this issue, in case someone else is looking for it: https://github.com/faye/websocket-driver-node/issues/34

Some package lingering in the npm registry for longer than it should (like a whole extra week).

1 Like