1

I'm attempting to properly set up between the Deploy Module between these two by follow these instructions. I've used Drush QuickDrupal to set up two new Drupal 7 instances.

$ drush qd d7-deploy-stage
$ drush qd d7-deploy-live

Initially, I ran them as on the same host machine via PHP's internal development server

$ php -S localhost:8000 # d7-deploy-stage
$ php -S localhost:9000 # d7-deploy-live

The end point is listed as http://localhost:9000/services/rest and the authentication (and only user) is admin with password of admin.

However upon a cron run on the d7-deploy-stage instance I

DeployAuthenticationException: Authentication error:-111 Connection refused in DeployAuthenticatorSession->deploy()

Question(s): How would I begin to debug this? Since the error reads "connection refused", I'm also curious if I start the servers with a fully qualified domain name (or my machine's IP) or if the port numbers an issue in the addressing

What I've tried so far:

The Patch: The '#31' has failed with the following output patch -p1 < deploy-404_CSRF_error-2017767-31.patch

patching file plugins/DeployAuthenticatorSession.inc Hunk #1 FAILED at 76. 1 out of 1 hunk FAILED -- saving rejects to file plugins/DeployAuthenticatorSession.inc.rej

Update

On both instances, based this answer below, I downloaded the 'add start up .ht.route' patch, applied and restarted

wget https://www.drupal.org/files/issues/add_a_startup-1543858-58.patch
patch < add_a_startup-1543858-58.patch to each instance.
php -S public-ip-number:8000 .ht.router.php # stage
php -S public-ip-number:9000 .ht.router.php # live 

The error on drush cron has changed

WD deploy: DeployAuthenticationException: Authentication error: 404 Not found : Could not find resource user. in DeployAuthenticatorSession deploy() (line 76 modules/deploy/plugins/DeployAuthenticatorSession.inc).

2
  • While that is a similar question, I've tried the patch with no luck. My question also asks a broader question in which I hope to see if my problem is inherit in the module or the testing infrastructure. That is, is my end point localhost with port number a bad idea? Is a FDQN a requirement?
    – Rick
    Commented May 24, 2016 at 16:36

1 Answer 1

1

There are few issues with your examples.

  1. You're missing routing-script for the built-in PHP web server. It's recommended to use since you're using PHP built-in web server and by default it may have some problems handling certain clean URLs (for example URI containing dots are treated as direct files). So you should apply this patch from #1543858 and re-run you server as below (same file you can use for both Drupal 7 & 8):

    php -S localhost:8000 .ht.router.php
    
  2. You should use -l in order to specify URI of the Drupal site to use, so your code knows what's your proper hostname (since the fail is related to the connectivity issue).

    drush -v -l http://localhost:8000/ cron
    

If you still have Connection refused, you should check where the code is trying to connect and check the connection (e.g. wrong configuration or firewall in place). To debug the connectivity issue, on OS X you can use dtruss, on Linux this can be debugged by strace, e.g.

strace -e trace=network -fp $(pgrep -fn php)

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.