How to Salvage Your Web Site When Your Web Host Deletes It

I am still stinging from last week’s web tragedy, but I am amazed at how fast the recovery process is moving. I thought I would share the steps I took to restore my site. Let me say up front that I DID have backups of my files, but I made the mistake of storing them with my web host who somehow managed to “lose” them in the process of overwriting every file on my site. No joke. It was a complete nightmare! I hadn’t ever downloaded any of those backups because they were so huge. (I have since purchased a terabyte wireless drive from Apple and it is backing up my Macbook Pro as I type this).

So how does one go about recovering from the complete obliteration of 3 years worth of work?

1. First you cry. :(( Yeah – let it out. Trust me that helps.

2. In my case, the directory structure was still in tact, but the files had all been overwritten. I downloaded what was left so I could rebuild the directories in the exact formation. This was a valuable reference for me.

3. Round up all local copies you may have of previous backups, images or code and stick them in a folder for quick reference.

4. Buy a copy of Juicer (it’s a Mac app, but there are Windows equivalents). Well worth the $17 I paid for it. Juicer will extract images and files from your browser cache among other things. In my case, I had just done some major debugging to my main stylesheet which would have been preserved in the next round of backups that were slated to be done the day after the tragedy occurred. Rather than spend 12 hours trying to figure out what I changed to fix it (dang IE7 issues), I used juicer to find my most recent copy in Safari’s cache. Copy/Paste and voila – stylesheet recovered. This also worked for the majority of recent images I needed. Many times I create a quickie image in Photoshop but don’t bother to save a copy after I upload it to the post.

5. My database was not lost, but it was mangled a bit. The apostrophes and quotation marks were all replaced by strange characters. Fortunately, I found the search & replace wordpress plugin that fixed my entire database in about 5 minutes. :D

6. My images are another story. I am slowly piecing them back together, mostly from my browser cache and searching by file name for local copies. (A mirrored replica of this directory would have been a lifesaver. I have older backups so at least I’m not starting from scratch here.)

7. For the truly mangled pages (there were a few), I used Google cached pages and Alexa cached pages to grab the text and images (if available) to once again copy/paste and reassemble the page. You can also used these cached pages to view source code and rebuild php templates or grab code snippets. In my case, I had just added a new form that hadn’t made the backup round yet. I was able to grab the code from a cached Google page and paste it back into my template. ;)

8. Check your 404 log. Once I have the majority of my site back up, I will be using the wordpress 404 plugin to send the log to my RSS reader. By checking this, I will be able to see anything I might have overlooked and get it fixed.

In total, since I started rebuilding, I am guessing I have less than 20 hours into the recovery process and all I have left to do is restore a handful of images and the occasional garbled page. Not bad considering I thought the world had ended last Wednesday. :D