I have free tier ec2 (1GB RAB, 500MB swap memory)
When Jenkins performs npm install on the project, ec2 just hangs, stops responding and only reboot brings it to life.
With htop, I analysed the situation and figured out that RAM just exhausts and chokes and additional swap memory I created doesn't help.
Since I'm having a school project, I want to use free tier and somehow launch the project there (I tried npm run build too, but for now it has too many issues).
I tried yarn too, but there's an issue with different versions of babel-loader being required by two other packages.
Is there a way to perform npm install in an automatic fashion (so that Jenkins can do it with just a script shell) and without hanging ec2?
I ran into the same problem where the server just would stop installing. While it is definitely possible to upload the node_modules manually, I preferred to let npm do it for me. I solved it by deleting half of the dependencies before running npm install. After it finished i added them back and ran npm install a second time. Now everything is installed.
From now on npm install should work fine, since most dependencies are already installed. However adding or updating many dependencies at once might result in a hanging server again.
Related
I execute npm install under a vue project, and then it will get stuck when it installs 53455 files in node_modules.
No matter which package is being installed at this time, it stops as soon as the number of files reaches 53,455. This is the conclusion I have repeatedly observed.
I have been tossing for a whole day, and repeatedly clean the cache and reinstalling it, all useless.
My system is Windows 7 64 bit, nodejs is the latest version v14.15.3
It may fail due to some network error (or) to some admin rights. It's better to use administrator accessed cmd prompt.
then run npm install
I'm building a React Native app which, at this point, doesn't have too many dependencies.
Every time I update an outdated npm package, I am GUARANTEED to run into bundling errors. It could be any package. I then spend hours trying different ways to clear npm cache, project cache, etc.
Does anyone else have this issue? Is there a "standard" set of commands I should run after upgrading ANY npm package? This is extremely frustrating!
This screen shot shows today's flavor of the issue:
BTW, I'm on Windows 10, using VS Code. My React Native version is 0.63.1, Node version is 14.7.0 and npm version is 6.14.7
UPDATE:
After a couple of hours of messing around and removing node_modules folder several times and issuing the following commands a few times, it's now started to work. Unfortunately, I can't put my finger on what fixed it. As I said, I did the same things a few times.
Here are the steps I repeated a few times -- though once should be enough:
Deleted the node_modules folder
In root folder, I ran npm cache clean --force
I then reinstalled all npm packages by running npm i
Went into cd android folder and ran ./gradlew clean
Then in project root, I ran npm start -- --reset-cache
Unfortunately, doing this the first time didn't fix the issue. It looks like I have to waste a couple of hours for things to get back to normal! Any idea how to make this problem go away for good?
I think this error is that you have not start the metro bundle. Try react-native start and then re-run the app
I have a problem with my app, after running react-native run-android, the procees stop with blank page app at get stuck at loading dependency graph done... nothing happens. Everything was ok and suddenly not work anymore.
I tried everything i found on the internet... still not working. Any ideas?
package.json
This is what i get..
Probably, it might be an issue with watchman. Please try to close all terminals and emulator, clean the project and give the build. It did worked with me when i came across the same situation.
When you’re unable to load a bundle, look at the packager logs or the error message displayed in the Expo client to see if it’s related to the packager. If so, you should try clearing the packager’s state to reduce the chance the bug is related to a stale cache or corrupt process.
These instructions are for macOS and Linux, but the general ideas apply to Windows as well.
Stop XDE/exp, which should also stop the packager. Check your list of running processes to ensure these processes are not running.
Delete node_modules in your project
If your project depends on other local projects (e.g. has a file:
URI in its dependencies), clear those local project’s node_modules
directories too for good measure even though it’s probably
unnecessary.
Clear your Yarn or npm cache, depending on which you’re using, with
yarn cache clean or npm cache clean
Run yarn or npm i to install your dependencies again
Run watchman watch-del-all to clear Watchman’s state
Kill the watchman daemon process
Delete the packager’s cache directory with rm -fr $TMPDIR/metro*
Start XDE or exp
With exp, run exp r -c for good measure
And just to be sure, force quit the Expo client on your phone or simulator and re-open it.
None of the above answers worked for me. I had this issue after transferring my data from my old mac to my new one. It turned out that the brew installed python was not properly transferred. So, basically I uninstalled and reinstalled watchman.
brew uninstall watchman python
brew install watchman
I then restarted the packager, and I was ready to go.
I am facing challenges while creating package for all the clientside components of my project in a system which does not have internet connection (for windows).
I have installed NPM for windows in the system. I need to manage the node_modules and run gulp commands in the system to create the package.
Since there is no internet connection in the box, I decided to copy the node_modules from my local box to the box which does not have internet connection.
For copy task, I am using msbuild script but somehow it is not working for me. Also, I see that when I am trying to copy the node_modules manual from one folder to another folder I am unable to copy.
Installed Node version : v0.12.2
NPM version: 2.7.4
Can anyone help me to provide any working sample to fix the above problem.
I think your approach is generally sound. You can build on the internet connected computer no problem! Copy files over with a usb stick if you like.
Here are some tips that might help solve your problems.
If the box is offline, develop on your local machine and deploy to the box
If possible do the gulp steps also on your local machine
Otherwise you need to run the local directory install of gulp. e.g.
node node_modules\gulp\bin\gulp.js build
because you won't be able to npm install -g gulp
Point 3 will work for other global npm_modules too btw
If you are facing long paths (> 256 characters) which can cause copy problems
Try using npm dedupe to remove duplicates
Or try copying to a shorter path e.g to c:\proj instead of to c:\very\long\path\proj
Or explicity install a dependency which has a long path
e.g
npm install deepdep#1.2.4
and then prune that folder from the original
rm node_modules\package\node_modules\package_with_too_many_nested_folders
Or Install the latest npm (v3.0 or higher) which solves this issue once and for all
e.g. npm install -g npm
which will build a much flatter hierarchy for your packages. Requires removing and reinstalling all packages.
Point 5 is a notorious issue on windows which is not a problem on linux because paths can be extremely long. (unless you are mounting a windows a windows directory from linux)
Personally I would go straight for the latest npm version but you have a number of ways of getting around the issue if this is not possible.
I have a Linux server that runs some virtual Machines using KVM. Host machine is a Debian derivate with kernel 2.6.32-32(-pve).
In roughly 75% of the times I run npm install, it hangs while installing modules. I ran it in silly debug level and it always hangs at the line
npm info postinstall [module name] (older npm versions)
or
npm verb unlock done using [...] (newer npm versions)
forever.
It happens in every project on the host machine and on the virtualized systems for every node and every npm version I have tested (a lot, including the newest ones (2.7.x atm)).
It also is not deterministic. Sometimes it works, sometimes not and most of the times it hangs on a different module.
On my dev machine (which is in the same network, running OS X 10.10) it works fine.
I am not behind any kind of proxy.
What could possibly be wrong here?
EDIT: For the time being I solved this problem by checking in all my dependencies as gzipped files using https://github.com/JamieMason/shrinkpack
I was having the same issue for several hours, and couldn't work out what the problem was. I tried re-installing everything a few times, manually installing individual modules, switching from x64 to x86 versions of NodeJS, etc, and got nowhere.
In the end, I changed networks, from WiFi over ADSL2+ [~2mbps] to 4G Hotspot [~20mbps] and it worked in 30 seconds. No idea why, since it didn't seem like a network speed issue, but it solved the problem.
Hope this helps someone else, too!
I had the same problem on our Jenkins slaves based on Ubuntu, and it was solved by upgrading npm.
Default npm installed by apt-get has version 1.3.10 now, which is very old (shown by npm --version). There are at least two ways to upgrade it:
Run sudo npm install -g npm command. If you want to install a specific version you can run append it at the end of command: sudo npm install -g npm#2.1.3.
Or you could add NodeSource repositories using these instructions and run usual sudo apt-get update.
I have the same problem on Windows. I deleted my npm-cache folder and reinstalled npm. Everything started working normally after that.
The npm-cache folder can be found at
<your-drive>\Users\<your-name>\AppData\Roaming\npm-cache
Remember to turn on 'View hidden files because the AppData folder by default is hidden
You can reinstall npm by using the command,
npm i npm -g
On Windows make sure you are in a native CMD (not in VSCODE or whatever IDE). Also, try increasing the max memory limit for node:
set NODE_OPTIONS=--max_old_space_size=8096
I did the same, switched my network from WiFi to 3G and it worked.
I'm experiencing the same problem, this is the issue on Github:
https://github.com/npm/npm/issues/7862
There are some useful tips that seemed to help a few people.
For me doing $ npm install -g node-gyp (as suggested here: http://gangmax.me/blog/2013/05/13/resolve-npm-update-node-gyp-hung-problem/) fixed the problem.
I had the same problem and just run
npm cache clean
And it works!
Had this same problem in Windows. The solution was to restart after installing node and npm and then run command prompt as administrator.
In my case, a system restart did the work.
I ran npm install --verbose which suggested some lock on a file in npm-cache. And I think the system released the file (which was under use) after the system restart. I am using windows.
Try cloning the project again and then run npm install to install the packages in your terminal.
I ran into this with npm v6, and it turned out I was just being impatient. I just needed to wait a little longer and it proceeded to the next step.
npm cache clean --force
works for me :)
=> this method working with me When npm block in installation Package for IONIC installation and ReactNative and another package npm.
you can change temporary :
npm config set prefix C:\Users[username]\AppData\Roaming\npm\node_modules2
change the Path in Environment Variables set C:\Users[username]\AppData\Roaming\npm\node_modules2
Run the your command to install your package .
open file explorer copy the link C:\Users[username]\AppData\Roaming\npm\node_modules
ok file yourpackage.CMD created another folder Created "node_modules2" in node_modules and contain your package folder.
copy your package file CMD to parent folder "npm"
copy your package folder to parent folder "node_modules"
now run npm config set prefix C:\Users[username]\AppData\Roaming\npm
change the Path in Environment Variables set C:\Users[username]\AppData\Roaming\npm
now package working correctly with Command line