Wiki is down again

[Constantine Zakharchenko]

cryptopp.com/wiki is "experiencing technical difficulties".

It's rather frustrating.

Why not upload offline version of wiki somewhere on github?

[Jeffrey Walton]

There are a few reasons. Some of them may be easy to overcome.

First, MediaWiki uses a markdown language that has to be rendered to HTML. Second, I do not know how to dump the wiki as a HTML website. Third, I don't know how to dump the wiki as a collection of marked-down pages.

I tried to solve (3) some time ago for backup purposes. I wanted the raw source files (and not a database). I could not figure out how to do it. (2) and (3) are hard problems for me because of the MySQL database and my lack of database skills.

Also, I do the mysldump as a backup. But I cringe when I think all the problems we are going to have trying to restore it in real life.

Jeff

[Jeffrey Walton]

On Tuesday, March 27, 2018 at 3:55:10 AM UTC-4, Constantine Zakharchenko wrote:

I think I tracked the problem down to the wrong PHP command. We needed to call the updated Red Hat PHP located at /opt/rh/rh-php71/root/usr/bin :

    # tail -n 12 /var/www/html/update-wiki.sh

    # Always run update script per https://www.mediawiki.org/wiki/Manual:Update.php
    /opt/rh/rh-php71/root/usr/bin/php "$WIKI_DIR/maintenance/update.php" --quick

    echo "Restarting Apache service"
    if ! systemctl restart httpd24-httpd.service; then
        sleep 3
        systemctl stop httpd24-httpd.service
        systemctl start httpd24-httpd.service
    fi

    cd "$TOP_DIR"

But previously the script looked like this:

     php "$WIKI_DIR/maintenance/update.php" --quick

It would succeed most of the time, but fail on occasion.

I think it is poor engineering on Red Hat's part. The latest Red Hat and CentOS should come with a modern PHP and Apache. People who want the ancient stuff should have to do something special. Instead, Red Hat turned it on its head and make most of the world jump through hoops hoops for PHP 7 and friends.

Jeff