amp config failed to read :7979

[amp-cli]

amp config failed to read :7979

[AbdealiLoKo]

Note that I can see canary_81ukt has been created in my phpmyadmin

[AbdealiLoKo]

@totten This is regarding your comment - #35 (comment)

Would the table canary_81ukt be created if mysql was using local sockets ?

[totten]

A few things:

1. The general operation of the canary is to setup a demo app (db+http vhost) -- then send an HTTP request. The error indicates that the HTTP request failed (PHP Warning: file_get_contents(http://localhost:7979/index.php): failed to open stream: HTTP request failed! HTTP/1.0 500 Internal Server Error).
2. Did you try the last suggestion ("Tip: Try running "amp config" and/or restarting the webserver.")? amp test almost always fails on the first try because it doesn't know how to restart apache. But if you alternate the command (amp test => apache2ctl restart => amp test), it should get around this.
3. If it still doesn't work, inspect the Apache+PHP logs. Worst case, to figure out what's wrong, you can alternately send requests directly to http://localhost:7979/index.php and hack the files in ~/.amp/apache.d and ~/.amp/canary.
4. Aside: I find worldWritable (aka chmod 1777) to be problematic with many web apps. It'll work for a while, but eventually there are conflicts where the web-user and shell-user fail to manipulate each other's files. I really recommend linuxAcl or osxAcl. (Related: http://civicrm.stackexchange.com/questions/2923/civicrm-does-not-have-permission-to-write-temp-files/2937#2937 )

[totten]

Aside: When I run amp test locally, I'm hitting another error due to the recent change to use amp.phar. Will post when that's patched.

[totten]

OK, the autoloading bug that affects amp test when using PHAR should be fixed (#39). That wasn't your problem, but it touches related code.

[AbdealiLoKo]

Apparently that is what caused my problem (The logs said so)
Closing this now :)