boldgrid / boldgrid-backup Goto Github PK

Is your feature request related to a problem? Please describe.
If auto-restore is enabled (Site Check or other), and the last full backup is too old, then a user may not want is used.

Describe the solution you'd like
A setting can be added to specify the maximum age of a backup archive that can be used for an auto-restoration. A default value of 1 week may be ok.

Describe alternatives you've considered
Alternatively, we can choose an arbitrary length of time and not have a user setting.

• Update the "Backup Complete" notice to show "with errors".
• Update the "Archive Details Page" and show if any errors occurred.

Maybe use something like this - https://stackoverflow.com/questions/2885468/getting-ftp-put-progress

Having WordPress < 5.0, go to BoldGrid Backup >> Settings and click to configure Web Server. An update notice is at the top of the modal.

Describe the solution you'd like
Add relevant PHP settings such as memory limit, max upload size, PHP version, max exec time to pre-flight check screen for diagnostic purposes.

We were unable to create a backup of your website due to the following:

Fatal error: Uncaught Error: Call to a member function add() on null in /wp-content/plugins/boldgrid-backup/admin/class-boldgrid-backup-admin-core.php:1874

Stack trace:

#0 /wp-content/plugins/boldgrid-backup/admin/class-boldgrid-backup-admin-core.php(2482): Boldgrid_Backup_Admin_Core->archive_files(true)

#1 /wp-includes/class-wp-hook.php(286): Boldgrid_Backup_Admin_Core->boldgrid_backup_now_callback('')

#2 /wp-includes/class-wp-hook.php(310): WP_Hook->apply_filters('', Array)

#3 /wp-includes/plugin.php(465): WP_Hook->do_action(Array)

#4 /wp-admin/admin-ajax.php(167): do_action('wp_ajax_boldgri...')

Describe the bug
Selecting Backup Site Now shows that the backup completes instantly, but the backup is still running in the background.

To Reproduce
This behavior was only observed on websites that generate very large backups.

1. Go to Backup Archives.
2. Click on Backup Site Now.
3. See the message stating Backup completed successfully.
4. Open the file manager for the website
5. Navigate to the backup directory.
6. See temporary backup .zip file.
7. File size increments when refreshing the view.

Expected behavior
The progress bar should display at all times while a backup is generating.

Desktop (please complete the following information):

• OS: Fedora 29
• Browser: Firefox
• Version: 65.0.1

If the user is really enjoying the backup plugin, let's ask them to leave a review.

You have 3 plugins and 1 theme that need updates, *click here* to create a rollback archive and update

Please reference - spatie/laravel-backup#762

This error occurs when any file added to the archive ( $zip->add(...) ) was deleted before calling the $zip->close();.

When working on this ticket, I think first step may be to try and replicate this issue (what an obvious statement, right?). IE, mid backup, delete a file that was already added to the file.

PHP 5.4 is the minimum.

Describe the bug
When an automatic backup is made before an auto-update two email messages are sent.

To Reproduce
Steps to reproduce the behavior:

1. Check your settings: notification email address and auto-backup before auto-update is enabled.
2. Set a plugin to auto-update.
3. Change the plugin version number.
4. Browse the site or admin pages (to trigger the auto-update).
5. Check your email.

Expected behavior
Only one notification email should be sent.

How does / how should BoldGrid Backup work with multisite?

Preliminary testing shows the following:

All backups are visible on ALL sites
You would have to name each backup the name of the site or something to tell them apart
Restoring site1 backup from the dashboard of site2 restores site1 as expected, nothing happens to site2

From Todd:

The email - that needs a bit more work. It should contain more about how to check to be sure the error is real and then specific advice on how to restore it including any precreated links we can supply.

Your Web Server backups are using over 10G, *click here* to configure retention settings

Description
Plugin attempts to make an FTP connection with every page load within the plugin. Viewing Settings, Tools, Preflight Check, or any of the other settings causes the plugin to attempt an FTP login.

When FTP option is setup user has to wait for the connection to timeout before the plugin page loads, and requires killing the processes server side to bypass waiting for the connection to timeout.

How to restore a backup if WordPress isn't working? Create a script to restore a backup.

When running bgbkup-cli.php restore method=pclzip or any other method except for "ajax", the logic chooses a non-ajax method based on availability instead of using your specified choice.

When an auto-update backup is in-progress, go to the WordPress Updates or Plugins page; the admin notice shows the progress completes ("Backup complete"), and another notice saying "Update protection activated" "Backup created successfully", but after performing an update (such as for WordPress Core), there is no rollback notice. In this scenario, the update protection should not be activated, and there will be no auto-rollback.

Auto-updates do not trigger the rollback scenario. If a (full) backup was recently made, and anything is updated manually withing a short time-frame, the rollback cron and notice should be in effect.

Note: In the future, we may have auto-rollback after something goes awry in an auto-update.

Describe the bug
When having a remote storage archive (such as on Google Drive) with no local/web server copy, downloading to the web server may take a long time. If a user refreshes the Archive Details page, it appears to have already been downloaded, but there is missing details and nothing in the archive browser. If the user clicks "Download to Local Machine", then the partial archive is downloaded. If it is then opened, it appears to be ok, until it is extracted, which results in a data error.

To Reproduce
Steps to reproduce the behavior:

1. Create a backup.
2. Upload to a remote storage location.
3. Delete the local/web server archive file.
4. Click Upload next to the remote storage type.
5. Refresh the page before the download completes.
6. Examine the Archive Browser and the archive details.
7. Click Download to Local Machine to examine the size and contents against the good copy in remote storage.

Expected behavior
If there was a checksum (such as MD5) stored in the archive log, then the archive details page could compare it against the file. If the checksum is not found or does not match the archive file, then the file is not shown to be local (on the web server), and no local controls are available.

Currently a notices pops up which to send the user to BG Central and give them a link. While we cannot install the plugin, we can provide a Download link , see: https://developer.wordpress.org/plugins/wordpress-org/detailed-plugin-guidelines/#8-plugins-may-not-send-executable-code-via-third-party-systems .

What I had in mind is creating an endpoint on the asset server that provides a zip file:
https://api.boldgrid.com/v1/plugins/boldgrid-backup-premium/download . API key authentication would be required to access the URL.

Is your feature request related to a problem? Please describe.
Using only whole numbers to display the size of backup in the progress bar leads to problems with users thinking their backup has stalled and is no longer working, especially with larger backups that take a significant amount of time to complete.

Describe the solution you'd like
Rather than showing something like 2 GB and then jumping to 3 GB after sufficient progress on the backup has been made, the progress bar should at least display increments like 2.3, 2.56 etc.

Additional context
Generally this is not a problem as backups on smaller websites complete within a few minutes, but larger sites where the compressed backup size is 3+ GB and the backups can take 20+ minutes to complete this creates some confusion

Describe the solution you'd like
Being able to supply admin crendentials and a URL in the origin website interface could help simplify the transfer/migration process.

The list of backup archives is ordered by descending date/time for the local "Web Server" storage types, but not for others; they appears unordered.

To Reproduce
Steps to reproduce the behavior:

1. Create some backups.
2. Upload some backups to other storage locations.
3. Delete some of the local backups, so that they exist only in remote locations.
4. Examine the list.

Expected behavior
The list should be ordered by date/time for all types.

src=bgbkup-nav

[14-May-2019 19:02:22 UTC] PHP Warning:  fread() expects parameter 1 to be resource, boolean given in /home/user/public_html/wp-content/plugins/boldgrid-backup-premium/admin/remote/google_drive_archive.php on line 283
[14-May-2019 19:02:22 UTC] PHP Warning:  feof() expects parameter 1 to be resource, boolean given in /home/user/public_html/wp-content/plugins/boldgrid-backup-premium/admin/remote/google_drive_archive.php on line 278

What happens with the filepath does not exist?
boldgrid_backup_jobs

1 => 
  array (
    'filepath' => '/home/user/boldgrid_backup/backup.zip',
    'action' => 'boldgrid_backup_google_drive_upload_post_archive',
    'action_data' => '/home/user/boldgrid_backup/backup.zip',
    'action_title' => 'Upload backup file to Google Drive',
    'status' => 'pending',
    'start_time' => 1557862503,
  ),

I deleted the readme, did the restore, and I did not have my readme back.

boldgrid-backup/cron$ ls -alh ../ |grep readme
-rw-r--r--   1 bradm bradm  17K Apr  2 13:59 readme.txt

boldgrid-backup/cron$ rm ../readme.txt 
rm: remove regular file â../readme.txtâ? y

boldgrid-backup/cron$ ls -alh ../ |grep readme

boldgrid-backup/cron$ php bgbkup-cli.php restore method=ajax
Starting restoration process...
Attempting to restore "http://domain.com/single-site-1" from backup archive file "/home/user/boldgrid_backup/boldgrid-backup-domain-12345678-20190402-174753.zip"...
Using Ajax URL address restoration process...
Success.

boldgrid-backup/cron$ ls -alh ../ |grep readme

Please see attached image.

Other restore methods seem to work, just not ajax.

• Add a "Start Here" link.
• Add a "Settings" link.
• Add a "Get Premium" link.

Here's an image for reference:

My site just crashed. Have a cron that will figure this out and auto restore site.

Describe the solution you'd like
Create an option to configure a remote FTP path for backup storage during setup and allow users to create subfolders so they can store backups with a single user across multiple sites.

Describe the bug
The 4 tabs (Backups/Settings/Preflight Check/Tools) work as expected but the navigation and the design panels in each tab are missing.

To Reproduce

1. Upload and Activate Backup 1.9rc
2. Enter Key
3. Dashboard > BG Backup > Settings > Settings
4. The navigation on the left isn't clickable and there's no design panels on the right
5. Same with 'Get Premium' notice on Backup tab and the 2 links on Tools

Screenshots

Describe the bug
Some hosts make WordPress Core files read-only (root:root 444), which makes restorations fail if they include core files.

To Reproduce
Steps to reproduce the behavior:

1. Use such as hosting account or change the ownership of some core files to restrict the current user.
2. Restore a backup.
3. See the error.

Expected behavior
Write failures should be ignored and logged, the restoration process completes, and the errors are presented after the restoration.

Describe the bug
URLs need to be escaped, like this one (notice the &section= param):

<a href="admin.php?page=boldgrid-backup-tools§ion=section_locations">click here</a>

• Ensure Google Drive is listed alongside Amazon S3
• Ensure logos are used for Amazon S3 and Google Drive
• Ensure premium storage providers are explicitly listed on Backup Archive Details page.

Reference to Backup Archive Details page:

Is your feature request related to a problem? Please describe.
When generating a custom backup it can be difficult to determine the total size of the files/directories that are being backed up.

Describe the solution you'd like
Add a rudimentary file browser with disk usage statistics to inform the user what paths should be included or excluded in custom backups.

FTPES is currently taking a bit longer than FTP, example:

The "Backup Archives" page is taking much longer to load.

FTPES
24s
22s
23s

FTP
7s
7s
9s

Adding a passive / auto detect option would help speed things up.

Is your feature request related to a problem? Please describe.
Servers that are set up to force High encryption requirements require modifications to class-boldgrid-backup-admin-ftp.php to work correctly. All instances of 'ftp_connect' in the file need to be changed to 'ftp_ssl_connect' to work correctly.

Describe the solution you'd like
Having an option in the plugin settings would make this easier for users to configure.