Generally you would use phpMyAdmin within the Plesk Control Panel to perform a backup of your MySQL database ( Exporting and importing a MySQL database ) but sometimes this process can fail or time out if you have a reasonably large database. The work around for this is to use a simple PHP script to invoke a mysql dump command and place the resulting backup file into your private directory within your FTP area for download.

Solution

Create the following PHP file either directly from the Plesk File Manager or on your local computer and then upload it to the httpdocs directory. You could call it mysqldump.php

Copy the following line into your PHP file:

<?php 

set_time_limit(600);

system("mysqldump -h localhost -u YOURUSERNAME -pYOURPASSWORD YOURDATABSENAME > /usr/local/www/vhosts/YOURDOMAIN.CO.NZ/private/BACKUPFILENAME.sql");

?>

Replace YOURUSERNAME with your database username, YOURPASSWORD with your database password (note that you do NOT put a space between -p and YOURPASSWORD) and YOURDATABASENAME with your database name and specify a custom backup filename if you wish.

Now run the command by visiting the mysqldump.php URL in your browser. e.g. http://www.domain.co.nz/mysqldump.php - Be aware that it may take a moment to backup the database, so please ensure you only visit the URL once and allow it to complete. 

You should now have the resulting backup file ready for download via FTP in your private directory. Note that any files within your private directory contribute to  your overall disk quota so remember to remove the file once you have downloaded the backup. Best practice would be to also delete the mysqldump.php file from your hosting account because it contains your database login credentials and also to avoid any accidental execution in future.

When connecting to IIS7 using FTP you get the following error.

530 Valid Hostname is expected

Reason

When you set up name-based FTP services IIS7 wants the sitename defined with the username so that IIS7 can associate the username with the site they wish to log in to.

Solution A

If you want more than one FTP site on your VPS you will need to :

Configure all FTP clients such that :
Host: Mysite.com
User: Mysite.com|Username

Note the "|" that seperates the Sitename and the Username.

Solution B

If you only have one FTP site remove the "Host Name" defined in the "Bindings" property of the FTP site.

Step 1 - Identify Replaced Drive

1. Log in to your server and as root run the following command:
   
   # gmirror list
   
2. This will produce output similar to the following
   
   Geom name: gm0
   State: DEGRADED
   Components: 2
   Balance: round-robin
   Slice: 4096
   Flags: NONE
   GenID: 0
   SyncID: 3
   ID: 3594034660
   Providers:
   1. Name: mirror/gm0
      Mediasize: 159999999488 (149G)
      Sectorsize: 512
      Mode: r2w2e5
   Consumers:
   1. Name: ad4
      Mediasize: 160000000000 (149G)
      Sectorsize: 512
      Mode: r1w1e1
      State: ACTIVE
      Priority: 0
      Flags: DIRTY
      GenID: 0
      SyncID: 3
      ID: 4064102945
   

   In this example we can see the status of the mirror is degraded and the only active drive is 'ad4'.

3. To find out which drives are available and the corresponding device name run the atacontrol list command.
   # atacontrol list
   

   This will produce a list of drives with device names which are connected to your server.

   ATA channel 2:
       Master:  ad4 <WDC WD1602ABKS-18N8A0/02.03B04> SATA revision 2.x
       Slave:       no device present
   ATA channel 3:
       Master:  ad6 <WDC WD2502ABYS-02B7A0/02.03B03> SATA revision 2.x
       Slave:       no device present
   

   From this we can see that device '*ad6*' is the device name of the new drive we can use to rebuild the mirror. You should already be aware of the device name from alerts generated by your server when the drive failed; the replacement drive will show up with the same device name.

Step 2 - Rebuild Mirror

1. Now that we have verified device name of the new drive we can easily rebuild the array using the following commands:
   # gmirror forget gm0
   # gmirror insert gm0 <device>
   

   Where <device> is the device name of the new drive we determined above. In the example above <device> would be 'ad6'.

2. To verify the drive has been added to the mirror and the mirror is being rebuild run the '*gmirror list'* command.
   # gmirror list
   Geom name: gm0
   State: DEGRADED
   Components: 2
   Balance: round-robin
   Slice: 4096
   Flags: NONE
   GenID: 0
   SyncID: 3
   ID: 3594034660
   Providers:
   1. Name: mirror/gm0
      Mediasize: 159999999488 (149G)
      Sectorsize: 512
      Mode: r3w2e5
   Consumers:
   1. Name: ad4
      Mediasize: 160000000000 (149G)
      Sectorsize: 512
      Mode: r1w1e1
      State: ACTIVE
      Priority: 0
      Flags: NONE
      GenID: 0
      SyncID: 3
      ID: 4064102945
   2. Name: ad6
      Mediasize: 251059544064 (234G)
      Sectorsize: 512
      Mode: r1w1e1
      State: SYNCHRONIZING
      Priority: 0
      Flags: DIRTY, SYNCHRONIZING
      GenID: 0
      SyncID: 3
      Synchronized: 3%
      ID: 2247886846
   

   We can see here the new drive has been added and the mirror is being rebuilt (3% complete). Rebuild times can take anywhere from 20 minutes to a few hours depending on the size of the drives and the amount of application I/O on your system.
   
   The first few lines show that overall state of the mirror, once the state changes from 'DEGRADED' to 'COMPLETE' the rebuild is complete.
   

Issue

RDP into Windows VPS fails with the following error:
 

Solution

Parallels Plesk Panel 10 Documentation

Parallels Plesk Panel Video Tutorials

By default, when a search engine spider visits your web site it follows all variable links, queries, search boxes and dynamic pages on your web site This can end up in an endless loop causing excessive load on your web site and generating high traffic.

Solution

The problem can be solved by uploading a robots.txt file to your web site which will prevent the search engines from following any URLs that begin with /? (these are the common dynamic page references).

Insert this text into your robots.txt file:

User-agent: *
Allow: /*?$
Disallow: /*?

Windows

1. Install the new Ironstor CBM
2. Next uninstall the new version to remove all legacy settings: 

    Open [Start] -> [Control Panel] -> [Add/Remove Programs]
    Select [Ironstor Cloud Backup Manager] from the list and press the [Remove] button

3. Repeat the installation of Ironstor CBM

MacOS

1. Remove all program files by opening a terminal session and typing the following commands. (You can open Terminal by typing 'Terminal' into Spotlight)

2. Remove all backup setting by removing ~/.obm

3. Restart your computer

4. Install the new Ironstore CBM

You wish to use CURL to get information from secure https:// sites but get validation errors. 

Cause

The CURL libraries no longer include a list of certificates to validate against.

As such, these need to be obtained and then referenced in your code.

An example is given below.

Resolution

Obtain CA Certificate bundle in PEM format from http://curl.haxx.se/ca/cacert.pem and save this to a location within the httpdocs directory structure.

Within your code reference this.

An example :