For the past three weeks, I have each day downloaded my website's access log by hand, at first literally, using FTP, and then, after a couple of days, with a command to do so. And each time, I would go look at the file and change it's name from access.log.0 to access.log.yyyy-mm-dd where dd is actually yesterday's date. The point of all this is to try and get into a position to make use of a neat-sounding analytical tool called Bise (see Seeking satisfaction in statistics).

I'm nowhere near that, yet, but storing the log files here is crucial. Soon, I'll be travelling without my laptop for a while, and I don't want to lose the files as they are rotated off the server, so it was time to automate the process. I ended up with a very simple script that first downloads and then renames the file. The cool bit was finding out that it is relatively easy to get yesterday's date in bash. $( date -v-1d +%F ) gives me exactly what I need.

Tomorrow I'll run the script one more time to check and then add it to my cron jobs. And while I'm about it, I might just add the other two backups that I do by hand most days.

Reactions from around the web


Webmentions allow conversations across the web, based on a web standard. They are a powerful building block for the decentralized social web.

“Ordinary” comments

These are not webmentions, but ordinary old-fashioned comments left by using the form below.