When you put a Drupal site in maintenance mode, non-administrators see the standard maintenance mode page (assuming you clear caches after doing so). If you examine the response, you will see that it is sent back with a HTTP status code of 503, which from RFC 2616 is:
503 Service Unavailable
The server is currently unable to handle the request due to a ...
Installing sites in sub-directories is supported in Aegir 2+ without any such hacking. Just enable the "Site subdirectories" feature, under the 'experimental' group at admin/hosting/features.
That said, to answer your question, DB credentials are available via drush_get_context('site').
Yes there is: it's a two or three step process, but automatable.
Step 1) Create a Drush alias, or in Aegir terminology, 'context' for the site using drush provision-save (arguments)
Step 2) Install the site using drush @(site context name goes here) provision-install
Step 3) (optional but recommended) Verify the platform in the Aegir interface to have it ...
Unless you've done something custom to allow it, Google can't crawl your site in maintenance mode.
Because you need to be logged in to view, googlebot will see the designated maintenance page.
For additional guidelines from Google see the following links:
Webmaster Guidelines: https://support.google.com/webmasters/answer/35769?hl=en
SEO Guidelines (multi-...
You can follow the example from the profiles/standard/standard.install file and use code like this inside your implementation of hook_install():
// Create a Home link in the main menu.
$item = array(
'link_title' => st('Home'),
'link_path' => '<front>',
'menu_name' => 'main-menu',
Edit /var/aegir/config/includes/global.inc and put in it:
$conf['file_directory_path'] = '/var/nasmount/<?php print $this->uri ?>/files';
$conf['file_directory_temp'] = '/var/nasmount/<?php print $this->uri ?>/files/tmp';
This will be applied globally to all sites. If you only want to make this change for one or two sites, create ...
The answer is to mv siltes/firstsite.com/files to /var/nasmount, then create a symbolic link to it.
(from inside your sites/firstsites.com directory, mkdir /var/nasmount/firstsite.com, mv ./files /var/nasmount/firstsite.com, ln -s /var/nasmount/firstsite.com/files files).
I would put this in a script so this becomes a drush command, then make it run at the ...
I would recommend that you avoid multisite, and clone the code into a separate codebase -- then copy the database. This is much easier to manage for simple use cases, and would entirely avoid your problem above, as you would only have one copy of each module present in your code base, so Drupal would not be able to load any of them twice.
I am not sure ...
Our current (simplified) practice at work:
Pull down the drupal core git repo http://drupalcode.org/project/drupal.git/
Add upstream to your local project that tracks the above repo
When a new version of drupal is released, fetch the drupal repo and merge the appropriate tag
git fetch upstream
git merge 7.7
Here is an issue on Drupal Groups: http://groups.drupal.org/node/96639.
Check if your system is Ubuntu (the installation script assumes that
Remove everything in /var/aegir and run the installation script again.
Run 'mysql_secure_installation' command and finish it successfully. Aegir
checks for security flaws during installation ...
Is this error displaying on the site's front-end? Or in your apache logs? And are you planning to run many more OpenAtrium instances?
If you're getting that error on the front-end, and aren't planning to run very many more OA installs, just up the memory_limit in php.ini, as it looks like you've actually got RAM to spare, and individual processes are ...
Have had similar issues with this in the past,
How many instances of drush is there on your server? In my case I had a drush 3.3 version in /usr/share/drush and the aegir 4.4 version (with provision and hostmaster commands) in /var/aegir/.drush
In order to get drush to work, I had to alter the path for the aegir user to include /var/aegir/.drush first
There are two ways to install Aegir, the (recommended) automated way using a debian package or the manual (though mostly automated still) method. Both methods take care of downloading and installing Drupal 6 and all needed contrib modules and other libraries.
Sounds like you might be looking for the Domain Access module
The Domain Access project is a suite of modules that provide tools for running a group of affiliated sites from one Drupal installation and a single shared database.
The cause of above error is that @sites is returning no sites.
Try drush sa @sites too, and see what it says.
Note, however, that @sites refers to all of the sites in a multisite install; it does not refer to all of the sites in your aliases.drushrc.php. If you rename your aliases file to foo.aliases.drushrc.php, then @foo will reference all of the ...
You can try drush archive-dump default,example.com,foo.com command to archive the DB and codebase. If it is not possibel through it, then you can create a shell script, which creates backup of the codebase and database.
It almost looks like the symbolic link they have you create in step 3.1.1, is not working correctly and may be causing this issue.
I don't know anything about the differences between Debian and RH 5.5, but maybe RE doesn't have a /etc/apache2/conf.d directory, in which case you would have to manually add an include line into httpd.conf
If you edit edit /var/aegir/config/includes/global.inc, as mig5 is proposing, and the sites are running on remote servers (not on Aegir master), then you should additionally verify the remote servers in Aegir, in order to get the changes in /config synchronised on remote servers.
Check out item 2 on this page on setting up sites in Aegir.
Login to your DNS control panel at your domain registrar, or wherever
your nameservers are, and setup the DNS for the domain name to point
at your Aegir server. It's worth including a wildcard record - ie an
'A' record for '*' to resolve to your aegir server - then you can
easily setup ...
The whole server doesn't restart when a site is added, only Apache, and it's just a graceful restart, so no real downtime..
Also, if what you mean by 'subsite' is http://example.com/subsite1, /subsite2, /subsite3, Aegir doesn't support that at this stage. Subdomains will work fine, but not 'subfolder' sites. See http://community.aegirproject.org/faq#...
Aegir is a powerful tool to manage hosting environments, deploy new versions of your site, clone a live site to a testing environment... Stuff like that. I don't know which kind of users you're talking about, but I would only give access to experienced Drupal developers. If you want 'normal' end users of a site to be able to setup their own subsite, maybe ...
I figured out a work around, actually it is quite simple. First ssh in as root. Then switch users su - aegir. Now drush will work. then simply use the drush command with the @sitename, so for instance drush @sitename.com dl themename. now this downloads to sites/all/modules, if you want you can move it to sites/sitname/modules. Either way it's a pretty fast ...
Here's the recommended way for Barracuda/Aegir by Omega8:
The recommended (and also more advanced) way to manage your Sites code
is to use Makefiles (and then Drush Make) to build your Platforms,
apply patches to contrib or custom modules etc, then using Profiler
module to create your custom Installation Profile and manage in your
Git repo just ...
If you do not know whether maintenance mode is blocking "everything you want blocked" by 100% (!) -- but at the same time are very concerned about "whether someone like google might still access it".
... then maintenance mode might be a bad choice for blocking your development site in the first place.
Personally, I recommend ...
The problem turned out to be the PHP version on the server. The Rest Server had a line of code that requires PHP 5.3.
There is an issue at https://drupal.org/node/2003396 and a patch at https://drupal.org/files/ServicesRESTServerFactory.inc_.patch
Installing the patch solved the problem without the need to upgrade the php version on the server