See the errors below. Does the npm install wipe out the files under /tmp? My downloads usually goto /tmp and they are wiped out. Why does it need to clean out the /tmp dir?
Thanks
$ curl http://npmjs.org/install.sh | sh % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 3902 100 3902 0 0 16745 0 --:--:-- --:--:-- --:--:-- 84826 fetching: http://registry.npmjs.org/npm/-/npm-1.0.15.tgz 0.5.0 1.0.15 prefix=/usr/local/node altprefix=/usr/local/node-v0.5.0
This script will find and eliminate any shims, symbolic links, and other cruft that was installed by npm 0.x.
Is this OK? enter 'yes' or 'no' yes
All clean! ! [ -d .git ] || git submodule update --init node cli.js cache clean node cli.js rm npm -g -f --loglevel error shell-init: error retrieving current directory: getcwd: cannot access parent directories: No such file or directory
node.js:360 var cwd = process.cwd(); ^ Error: No such file or directory at Function.resolveArgv0 (node.js:360:23) at startup (node.js:48:13) at node.js:474:3 gmake: *** [uninstall] Error 11
node.js:360 var cwd = process.cwd(); ^ Error: No such file or directory at Function.resolveArgv0 (node.js:360:23) at startup (node.js:48:13) at node.js:474:3
node.js:360 var cwd = process.cwd(); ^ Error: No such file or directory at Function.resolveArgv0 (node.js:360:23) at startup (node.js:48:13) at node.js:474:3
node.js:360 var cwd = process.cwd(); ^ Error: No such file or directory at Function.resolveArgv0 (node.js:360:23) at startup (node.js:48:13) at node.js:474:3 It failed
It appears that I am the only one having this issue!
Is anyone able to successfully install npm under the latest node release (v0.5) on CentOS 5.5? Every time I try 'curl http://npmjs.org/install.sh | sh' as root, it empties my /tmp clean!
That's really odd. If you set something else as the TMPDIR environ, does it clear out that folder as well? Does it do this on other linuxes? Is there any other error output?
On Fri, Jul 8, 2011 at 16:42, S. A. <game...@gmail.com> wrote: > It appears that I am the only one having this issue!
> Is anyone able to successfully install npm under the latest node release > (v0.5) on CentOS 5.5? > Every time I try 'curl http://npmjs.org/install.sh | sh' as root, it empties > my /tmp clean!
This worked! folders and files under some-folder survived: $ ls -laF ~/npm-tmp total 16 drwxrwxr-x 3 root root 4096 Jul 11 15:12 ./ drwxr-x--- 25 root root 4096 Jul 11 15:12 ../ drwxrwxr-x 2 root root 4096 Jul 11 15:12 some-folder/
I am now able to do 'npm ls', but when I did 'npm update', it is kind of stuck there for the past 10 minutes. I see that connect, express, mongodb and mongoose dir are updated with time stamp 15:18, and date now is: $ date Mon Jul 11 15:28:47 EDT 2011 npm update is still running. Is it OK if I break it? Does it leave any updates in an unstable state?
Yeah, it should be fine to kill it. Even if modules are in a broken state, you can always re-install them. `npm cache clean` if the cache is messed up.
On Mon, Jul 11, 2011 at 12:30, S. A. <game...@gmail.com> wrote: > This worked! folders and files under some-folder survived: > $ ls -laF ~/npm-tmp > total 16 > drwxrwxr-x 3 root root 4096 Jul 11 15:12 ./ > drwxr-x--- 25 root root 4096 Jul 11 15:12 ../ > drwxrwxr-x 2 root root 4096 Jul 11 15:12 some-folder/
> I am now able to do 'npm ls', but when I did 'npm update', it is kind of > stuck there for the past 10 minutes. I see that connect, express, mongodb > and mongoose dir are updated with time stamp 15:18, and date now is: > $ date > Mon Jul 11 15:28:47 EDT 2011 > npm update is still running. Is it OK if I break it? Does it leave any > updates in an unstable state?
Can you please post the full output as a new issue on the github issues page, along with the output of `npm config ls`? This isn't related to the original post in this thread.