When I run npm install it seems to work fine until part way installing packages. It seems to have no problem with the first half, but then after a while it will fail to be able to reach other packages. I just get the repeating errors, eg:
npm http request GET https://registry.npmjs.org/react-hot-loader
npm info attempt registry request try #3 at 6:43:34 AM
npm http request GET https://registry.npmjs.org/react-tap-event-plugin
npm info attempt registry request try #3 at 6:43:34 AM
npm http request GET https://registry.npmjs.org/react-test-renderer
etc.
It will continue to do this for an hour and and then the install will fail.
The install breaks at a different package each time so I don't think it's a problem with a particular file.
I can access these files fine with my browser and curl.
My work has a firewall but this domain is whitelisted.
Would anyone know what I could do to get this to work or what could be causing it?
If it's about the timing problem you should find a speed solutions for npm install.
So you can try these faster command than npm install :
pnpm install %70 faster
or
npm install --no-audit 15% faster
or
npm install --prefer-offline --no-audit 15% faster
check this article for details : speeding up npm install
You can override the max and min timeout in ~/.npmrc.
// npm config ls -l
// add these 2 lines in ~/.npmrc
fetch-retry-maxtimeout = 6000000
fetch-retry-mintimeout = 1000000
If your internet connection is the problem, try increasing the timeout:
npm config set timeout 6000000
The value is a 32-bit int.
If it's still relevant or maybe for other people of interest: For me it helped, deleting the package.lock file and running npm cache clean --force.
It might not be your case, but I had issues with a package being hosted at github with the repo url being only with git protocol (port 9418 not usually open on firewall).
Once added that to the firewall I could npm install without issues.
You can view the repository url with:
$ npm view zone.js repository.url
git://github.com/angular/angular.git
I saw an answer earlier that can resolve your problem by overriding the max and min timeout in ~/.npmrc file, but some didn't understand how to do it.
First go to the nodejs folder - for me it's located at ( Y:\Program Files\nodejs )
Open the node_modules folder.
Then npm folder.
Select (.npmrc) file + right click and open it with any text editor you want.
Finally add these 2 lines and click Ctrl+S
fetch-retry-maxtimeout = 6000000
fetch-retry-mintimeout = 1000000
.npmrc file after modification
npm cache clean --force
npm install --force
It works fine.
If you are on windows, try running vscode as administrator, it worked for me, I tried npm config delete https-proxy , npm config delete proxy and , tried deleting node_modules, and package.lock.json and ran npm cache clean --force but at last ran vscode as adminitrator before deleting node_modules and package.lock.json, it worked
I solved the timeout issue by executing these commands:
rm package-lock.json
npm i
Related
Using NPM to install dependencies, including one that is stored in Verdaccio running locally.
Somehow NPM is stuck attempting to load the Verdaccio dependency from localhost even though the command to install from the corrected location has been run npm install --save --registry http://CORRECT-URL ...
Using rm -rf node_modules package-lock.json and removing the dependency from package.json, then running the install again - still DOES NOT WORK. Every time the install command completes, package-lock.json STILL uses localhost to resolve the registry.
Removing localhost from ~/.npmrc and running npm set registry http://CORRECT-URL also didn't help.
Please help. Where is NPM remembering localhost and insisting on using it???
UPDATE - Using npm install -ddd
Attempted to use npm install -ddd to see where npm is picking up the package. The output shows the CORRECT-URL, but when it is done, package-lock.json still lists localhost.
Note all of the following:
Removed all references to the PACKAGE from package.json and package-lock.json
Verified no references to localhost in either file
Removed node_modules/PACKAGE_DIR with rm -rf
Removed PACKAGE from node_modules/.package-lock.json
Searched for .npmrc and npmrc in the build tree and found two empty files:
node/node_modules/npm/docs/public/configuring-npm/npmrc
node/node_modules/npm/.npmrc
~/.npmrc has 2 entries for the CORRECT_URL in the following forms:
//:/:_authToken="<AUTH_TOKEN>"
registry=http://:/
Stumbled upon this post while trying to debug the exact same situation. What ended up solving this for me was to blow out package-lock, run npm rebuild, then npm install --registry . Hope this helps someone in the future!
I'm having issues with JFrog NPM private repositories.
In our CI we configure the npm private registry and then do npm install like this:
- npm_private_registry https://DOMAIN.jfrog.io/DOMAIN/api/npm/npm-virtual/ $TOKEN
- npm install
This code works perfectly fine when we use the node:12.13.1 Docker image.
Our problem is that as soon as we change the image to a newer one, let's say node:12.16.1 then the npm install command fails.
We then get the following errors:
$ npm install
(wd=/builds/DOMAIN/APP)
npm WARN tar ENOENT: no such file or directory, open '/builds/DOMAIN/APP/node_modules/.staging/react-icons-921e2ffa/go/package.json'
npm WARN tar ENOENT: no such file or directory, open '/builds/DOMAIN/APP/node_modules/.staging/intl-58a6f707/locale-data/jsonp/gsw-FR.js'
...
...
...
npm ERR! Response timeout while trying to fetch https://DOMAIN.jfrog.io/DOMAIN/api/npm/npm-virtual/next (over 30000ms)
We tried various approaches already, but haven't been able to figure anything out just yet.
Does anyone have an idea where to start?
The issue seems quite generic. How is it possible to get timeouts on one version of the node image and no issues on another?
Thank you!
Seems like this link https://github.com/npm/cli/issues/1151 offers some explanation. The timeout got hardcoded down to 30000 ms in npm 16.4.0 via npm-registry-fetch. This has now been reverted to not timeout again https://github.com/npm/npm-registry-fetch/commit/fc5d94c39ca218d78df77249ab3a6bf1d9ed9db1. This fix is in 6.14.5 of npm https://github.com/npm/cli/releases/tag/v6.14.5.
In terms of default npm with node versions this means, releases 12.16.2 and 12.16.3 of node contain this problem via npm 6.14.4.
Good afternoon,
I am trying to install react and react-redux onto my project, but I keep getting errors that I don't understand.
$ npx create-react-app learn-redux
npm ERR! code ENOLOCAL
npm ERR! Could not install from "Phelps\AppData\Roaming\npm-cache\_npx\13464" as it does not contain a package.json file.
npm ERR! A complete log of this run can be found in:
npm ERR! C:\Users\Taylor Phelps\AppData\Roaming\npm-cache\_logs\2020-02-08T20_44_17_568Z-debug.log
Install for [ 'create-react-app#latest' ] failed with code 1
Could not install from "Phelps\AppData\Roaming\npm-cache\_npx\13464" as it does not contain a package.json file.
I am reading the error, but the problem is...that folder does not exist on my computer and I've checked in the terminal for npm and npx and both are installed on my computer. Does anyone have any ideas on why this may be happening and how to fix it?
Thank you.
According to this Github issue npx has trouble running on Windows if the user has a space in their name and from the error message I'd guess that is the case here. One of the suggested fixes is to set your npm-cache path to a folder without a space in it:
npm config set cache C:\tmp\nodejs\npm-cache --global
Then try running your command again.
Install the project in another directory except C partition :
npx create-react-app learnredux
Then :
npm i react-redux --save
This should work.
The problem is the npm cache by the path of the error. Try running:
npm cache verify
Verify the contents of the cache folder, garbage collecting any
unneeded data, and verifying the integrity of the cache index and all
cached data.
(from https://docs.npmjs.com/cli-commands/cache.html)
I published a module to npm. When I uninstall it and try to reinstall it, this is always done from a cache ( it is done even if I'm not connected to the internet, I just get a warning ). I would like to try if my module installs correctly from the remote npm repository on a fresh npm registry.
I tried to remove Node.js and reinstall it, but it does not help. There is no "npm" in the list of my applications that I could remove, only Node.js is listed. This is on a Windows 10 machine.
Where is the npm registry located and how can I remove it entirely?
You have to clean your NPM cache and then clear your registry using the below commands.
npm cache clean
npm config delete registry
Thanks
npm config delete registry
This worked for me
I'm suffering from the infamous npm install (update) hanging problem.
So far I found following recipes:
cleaned the cache: npm cache clean
deleted npm_modules in my project
set the registry to plain http (http://registry.npmjs.org/): npm config set registry http://registry.npmjs.org/
used --loglevel=verbose flag with npm install
increased number of connections as described here:
Increasing the maximum number of tcp/ip connections in linux
Yet it still hangs. The position at which it hangs seems to be random. It can be
npm verb get saving gulp-traceur to /home/me/.npm/registry.npmjs.org/gulp-traceur/.cache.json; or when installing npm verb afterAdd /home/me/.npm/q/0.9.7/package/package.json the last package downloaded is really random.
The versions are:
npm info using npm#2.7.5
npm info using node#v0.12.0
So the question is if there is anything else I can do about it?
I don't know if you have the same problem as I did but I can't make a comment to your question because I have not enough reputation.
Today somebody found solution to my similar problem. You can check it here:
Sometimes you need to use --force command to make sure the cache is cleaned:
npm cache clean --force.
Maybe related to this issue.
I tried all the above. I use homebrew and had to uninstall node / npm.
This worked for me:
brew install nvm
nvm install node
Afterwards npm started working again for me.