Linux Mint Update and Pelican Recovery
OK, I admit it. I goofed.
Linux Mint 19, Tara came out at the end of June and I was very excited about doing the update. I used instructions the 'How to upgrade to Linux Mint 19' tutorial to perform the steps. As instructed, I used the Timeshift tool to make a backup and proceeded. Unfortunately, the upgrade did not go well and I ended up with a non-bootable Mint installation. At this point, I should have followed the advise in section C2 of the upgrade tutorial and ran Timeshift from the bootable USB drive that I was using to recover my back up files. But, I didn't.
I have most of my data synchronized through dropbox and I also do occassional backups to an external hard drive. So, I decided to reformat my hard drive and do a fresh linux mint installation without retrieving the Timeshift backup. The installation worked and I was back up and running.
I have done fresh installs previously and I like doing them. Other than my core applications, I don't make a great effort install programs I only use occassionally. I only install those non-core applications when the need to use them arises. The end result is a much cleaner, tidier system than before the upgrade.
So, I reset up dropbox, and it synchronized my files. I pulled my other data backup off of my USB stick. Good to go. Well, almost. There were two exceptions to recovering my data. First, I did not have a recent back up of the files that are used to build this blog with Pelican. The second (and more painful) is I lost data files from the Gramps genealogy program that I use. My most recent backup was four months old on the external harddrive.
I will save my thoughts on recovering my genealogy data for later. But, I wanted to go through my steps for getting blog back and going.
Recovering This Site.
The good news is that my web site was still up and running. So, it contained all of my articles. In this case, it is a good thing that I am not a prolific content producer. So, I came up with the following plan.
- Restore the articles that I had saved from 4 months ago.
- Using FTP, copy down the HTML files and images that make up the content of the site.
- Find a way to convert from HTML back to markdown.
- Check the site thoroughly to make sure it is back in order.
Restore from Backup
This was easy. I just copied the files from my back up disk back onto my hard drive.
During this process I realized that I had added a lot of new information to old articles such as tags and better links. I also cleaned up awkward sentences (I am doing grow my writing prowess). I would need to revisit every article on the site to make sure it was back in some kind of order.
Copy HTML from the Site
Using Filezilla, I was able to copy my site locally. I made a 'recover' folder on my hard drive to hold the files. The first benefit was that I restored images and screen shots that I had lost.
Convert HTML to Markdown
I found a web site where you can Convert HTML to Markdown. Using the HTML file name, I created a new markdown file. The tags, publish date, and other meta data is found inside of the HTML file. That information gets transferred to the markdown file. Pelican places the body of a blog post between
<Article></Article> tags. Copying and pasting the article information into the coversion web site yielded the body of the article.
This process allowed me to recreate new articles from the last four months. I used the same process to review the older articles and get them into decent shape.
Other Side Effects
I also lost some of my Pelican configuration information. The basics were not difficult to rebuild since I had gone through the process of converting this site from Wordpress to a Pelican statically generated site at the end of 2017. I did find that the latest versions of Pelican and the tag_cloud plugin required some additional settings.
I am back up and moving forward. That's a good thing. There also need to be changes in the frequency and way I do backups on my home computer.