WP 3.2.1. and backWPup 2.0.3 – broke AGAIN

can u send me a log?

just send it to your email (mail@______.de) which we’ve found on your page.

Thanks for the help

Cheers

I have send the answere.

I htink it was only the Backup to Folder setting.

Daniel, thanks for all the work!

Ok, here’s the clue:

In former versions the folder for the backups was predetermined. Now you can choose your folder for every job individually.

Obviously you MUST set the target-folder for every job in your job-settings. Otherwise the zip (tar, …) file gets lost in the temp directory.

Again, thanks to Daniel – it’s up an running now! YEAH

I am having this problem as well – with WordPress 3.2.1 and BackWPUP 2.0.3 when i setup a job, and hit ‘run now’, it just hangs forever. Here is a picture of what it looks like: https://screencast.com/t/xPnvH7CsnpL

can you tset the development version ?

Now it says:
Can not create temp folder: /var/lib/php5/.backwpup_796747168/
Normally this folder is available only for root, others can’t even read it’s content.

but it is decleard as temp dir for the system.
with version did you use? try dev sher i have made many changes.

I use:
Wordpress 3.2.1
BackWPup 3.0.5
@
Ubuntu 10.04 with Apache 2.2.14 and php 5.3.1 (from repositories)

Thanks a lot, now it works!

I would like to say few other things, I think they doesn’t require new forum thread.
1. There are no backups in “backups” screen in situation when they are stored to FTP server and are not stored in local folder. Is this behavior expected?
2. Can you add support for sending email using wp_mail() function? This can be useful because it makes possible customizations using tools like Configure SMTP plugin.
3. There is a bit corrected description for option to use cron service of backwpup.com (original one contains few misprints):
If you check this than the job schedule will submited to backwpup.com. Backwpup.com will call your blog wp-cron.php to execute jobs. Use this service only if your hoster doesn't provide a cron service or if your blog has few visitors. Please consider making a small donation for the plugin if you use this service. The service can be removed by me at any time without notification.
4. May be I can help with Russian translations? Feel free to write me to koroboru [at] yandex [dot] ru. (replace at and dot with @ and .).

to 1. i will test it.
to 2. No i can’t, beacuse the backup working without wordpress in background. But i will look with var the plugins set and is it easy to read them. can you give my an example plugin ?
to 3. thanks i have chend it.
to 4. translations ever welcome. send it to mail [at] backwpup [dot] com

to 1. cant find a problem at moment.

I had the same problem as GOWPTG and after a few hours digging in code I think I found the reason.

At some hosting companies, PHP can not access the own(!) host via the https:// filewrappers, even though allow_url_fopen is On and external hosts can be accessed.

On such servers the plugin fails to internally continue to another page on the same server via a wp_remote_post() call with full https://yourserver.com/… url.

Here a short php-testcase, note that display of php-warnings has to be switched on:
[code]echo file_get_contents('https://'.$_SERVER['SERVER_NAME'].'/');[/code]
If you see a warning like this, you can’t use the plugin for now:
Warning: file_get_contents(https://www.yourserver.com/) [function.file-get-contents]: failed to open stream: Connection refused in /var/www/yourserver/testcase.php on line 1

ahhh, nice to now can on that hosts connect with localhost ?

No. The result for ‘localhost’ is:
Warning: file_get_contents(https://localhost/) [function.file-get-contents]: failed to open stream: HTTP request failed! HTTP/1.1 404 Not Found in /var/www/qrcweb/htdocs/testcase.php on line 1

Actually ‘localhost’ can’t work, as with shared hosting the hostname is relevant to find the correct account.

Seems to be a general configuration issue with firewalls/internal routing/whatever with some hosting companies, e.g. my company uses load balancers for the shared hosting.

Maybe you can add an option in the plugin to use the “old way” without extra trigger (if I understand the changelog right), or did that cause too many problems? I only started to use your plugin 2 days ago…