I have my symfony project on my dev box machine and also on a hosted environment where i have only FTP and HTTP access (i have a phpmyadmin access to my prod DB). This hosted version is an alpha release that should be initialized with fixtures.
Problem is : to do that, i have to write/update .yml fixtures in local, insert them in local DB with the symfony task, go on my local PMA, generate a data export, go on prod PMA and import the data...
Is there any way to generate a .sql file from my local fixtures so that i can insert them directly through my prod PMA ?
Thanks
The easiest is probably for you to look at this website:
http://brentertainment.com/2010/02/15/run-a-symfony-task-from-your-model-or-action/
You may run tasks from the web, you might make it a bit smarter and allow to run tasks from the web. Also clearing the cache might be a useful thing to extend to your admin interface for example.
I think you can't "out of the box" in symfony.
BUT you may be able to use the Doctrine Profiler to log all the SQL requests done by the sfDoctrineDataLoadTask task class into a file.
Are you using the symfony doctrine:build --all command? When I do that, a schema.sql file is created within the data/sql directory.
Related
I've successfully created site using Umbraco now its time to upload it on hosting server..
i've searched and got one paid product for the same..and i dont want to use it.
Has any body tried developing Umbraco site on local and then uploading it on server?
If yes then please help me doing that.
First I run the umbraco install from a local IIS website. Then I setup my visual studio solution for that website (and my souce control). Then I work, until I reach a beta version, then I go through this process for deploying:
Ftp over to the remove website and copy the whole website (I actually use Beyond compare).
Connect to my local database with management studio and create a .bak file.
Upload the .bak file to the database server.
Restore that database
Review connection strings in web.config
Then I'm pretty much done.
Once I'm "live" and have content I don't want to lose, when I want to work on the website, I bring back the live database through a .bak file, then I make my changes. They often include DB changes since the schema is basically in the database. I note all the operations I do. Once my changes are ready I manually replicate the changes on the live site as I update the files.
This is very painfull but I tried solutions like courrier and other things like that and they are not reliable enough for production I find. Manually is the only risk free way I see for the moment.
Hope this helps.
Yes, that happens all the time. Use FTP to copy your local installation to your webserver, modify the web.config to point to the correct database and your website should be up-and-running.
I'm sure there are more elegant solutions with less clicks but here's how I do it on azure websites with sql, not sure what hosting/db you're using:
1) Create an empty db on azure with the same login and user as my local db.
2) Create an empty site on azure connected to my db.
3) Download the publishing profile.
4) Upload the db the first time with Sql Azure Migration Wizard.
5) Import the publishing profile into and upload the site with WebMatrix.
6) Thereafter I deploy the site and db with WebMatrix.
WebMatrix uses WebDeploy or FTP, you can use WebDeploy through IIS if you like, and FTP.
Does anyone know a SQL query that will purge a MediaWiki database of old revisions? My database has grown out of control, and I need to prune it to make it possible to download and manage.
I don't have shell access so, I need to do this with a SQL query.
I have tried the solution suggested here, but it doesn't work http://www.mediawiki.org/wiki/Extension_talk:SpecialDeleteOldRevisions2#Deleting_only_archived_revisions
Thanks for reading :)
Nicholas
As you, I don't have shell access to my MediaWiki. So I can't do a lot of things like maintenance.
Here is my solution : host your MediaWiki web site on your computer just to do your maintenance tasks
Backup your database
Backup your MediaWiki folder
Setup Apache (the web server) on your computer
Setup MySQL on your computer
Restore your MediaWiki database on your computer
Put your MediWiki folder on the Apache root folder
Finally run the maintenance task you want using shell. I suggest you the deleteOldRevisions script
After that, rebackup the folder and the database and restore them on the remote host
Use the Maintenance extension and run the relevant maintenance scripts with it. Direct database manipulation is pure madness, and using a local LAMP install as suggested by the other answer quite cumbersome.
Shell access is really required to properly run a MediaWiki but this is a common problem, please report your experience with the extension on the talk page or file a bug if you find any.
I'm using subversion to develop a website with a designer and I'm using a remote server to the Subversion database storage.
Now I need to make visible on the Apache(same machine that Subversion server) all the changes/commits that we make to the repository.
This is possible to do?
I have searched inside the repository files but nothing looks like the name of the files of the project(PHP Project).
Can someone give me a clue on how can I make visible the changes to the repository in a website? Is there a way to connect the changes that I do the the repository to the website?
Best Regards,
Assuming you can already browse the repository using HTTP, simply add something like Repos Style for a nice layout and folder/file history features.
You could shell out to svnadmin.
I want to set up a production and development environment for a wordpress website. The Wordpress site uses a lot of plugins. These plugins consist of php files which I upload to a directory within the main wordpress directory.
When activating and modifying the plugins, much of the configurations are stored in the mysql database. Developing on wordpress means much of my "code" and "changes" are stored in the database, as opposed to physical files that can be committed to and updated from a CVS repository. So instead of having a CVS keep track of my changes, I have to frequently and manually do mysqldumps of the database on production and import it into the development server.
Is there a better way to set up a development and production environment where much of the development changes are happening in the database?
Use CVS's hooks to perform the dump whenever you commit. I'm not familiar with how CVS does them, since I've only done it with SVN. However, from reading a bit of the manual, it seems pretty similar.
This section of the CVS manual explains what happens, script-wise, when you commit. Take a look at the section further down, "C.4.4.2 Keeping a checked out copy". You should be able to modify this example to put in your mysqldump command.
I would like to make a complete backup of my whole joomla 1.5 based site from time to time. How would this ideally be done? Are there any common pitfalls? Not that I only have ftp access to the hosting server. Is there a step by step tutorial somewhere? I am using latest Joomgallery and Kunena 1.0.9 (Legacy mode).
Maybe there is a good way to automate this?
There's two parts of the backup you have to worry about, the database and the files.
The first part is the database. It can be backed up using something like phpMyAdmin. If you don't have this available on your server already, it's not too hard to upload and get it going yourself. From there, you can just Export the entire database to a gzip file.
The second part is the code and uploaded files. The code base shouldn't change too often, so you could probably just make one backup of this. There's a number of ways. The simplest is to just download the entire folder via FTP, though if you're Linux, I'm sure someone will know a single command line to get all the changed files (rsync?).
The database is the main thing you have to worry about though: everything else should be able to be rebuilt just by reinstalling.
I think this: http://www.joomlapack.net/ is what you need. I use it myself and it works like a charm. Both for backups and for moving my Joomla installations from developer sites and to the real site.
get an FTP synchronisation tool and keep an up-to-date copy of your site locally. Then you could run the batch script
mysqldump -hhost -uuser -p%1 schema > C:\backup.sql
to create a backup of your mysql tables at various points in time.
edit
you would have to have MySQL Server installed on your local machine and path to its bin directory in you PATH, in order to run the mysqldump command without much hassle. -p%1 would take the command-line provided password, as you wouldn't want to store passwords in your batch script.
If you only have FTP access you are in a bit of a problem, as beside all files you'll also have to backup the database. Without accessing the database, a full-backup won't do you any good.
Whatever backup strategy you choose - be sure it can handle UTF-8 correctly. Joomla 1.5 stores all content with UTF-8, even when the database charset is set on 'iso-5589-1' - so when the backup solution is detecting the database charset, some characters like € or é will result in "strange" ¬ / é - not really what you'll want.
I absolutely endorse using Joomlapack - it works great. The optional remote tools allow you to initiate the backup from a Windows desktop machine - it performs the backup and downloads it. The remote has a scheduler, and you can also set it off to backup and download a list of sites.
Joomlapack also provides a file "kickstart.php" which you copy to your empty server account along with the backup, which automates the restore procedure. You do have to create an empty database with PHPMyAdmin or similar, and you are given the opportunity to supply the database parameters (host, database, username, password) during the process.
One pitfall I did run into with this though is that some common components can have absolute URLs in their configuration - e.g. SOBI2, Virtuemart. It's then just a matter of finding the appropriate configuration file, editing it and re-uploading it.
Another problem was one archive file (either ZIP or their JPA format) got a filename with a "?" character in it (from a Linux server) and this caused a bit of a problem trying to install it locally on a Windows WAMP stack - the extract process on the ZIP file failed, and it stopped the process completing cleanly.
I suggest using automatic backup service by http://www.everlive.net
Update:
Ok, here is some more information. EverLive.net is a website where you can create a free account. Enter your website details and you are ready to take your backups withe just one click. Restore is also possible in the same way.
Further you can use automatic backup option to take automatic backups at defined intervals. Other than that, you can use the website health check service to inform you if your website is not available.