https://blog.pgxn.org/tags/httpd/HTTPD2010-08-13T23:45:00ZThe PGXN MaintainersHugohttps://blog.pgxn.org/post/949562598Start Your Mirrors!2026-10-07T16:13:48Z2010-08-13T23:45:00ZDavid E. WheelerWork has finally started. We now have mirroring. This was the first task for the project, and it’s now checked off on the status page. Hurrah!

This turned out to be a pretty simple task, of course. For now I’m using the Kineticode (my other employer) server to host both the PGXN site and the master mirror. This is temporary, just a way to get things up and going as quickly as possible. The master mirror runs rsyncd with read-only access to the pgxn path. So all you have to do to mirror it is set up a cron job like:

]]>Work has finally started. We now have mirroring. This was the first task for the project, and it’s now checked off on the status page. Hurrah!

This turned out to be a pretty simple task, of course. For now I’m using the Kineticode (my other employer) server to host both the PGXN site and the master mirror. This is temporary, just a way to get things up and going as quickly as possible. The master mirror runs rsyncd with read-only access to the pgxn path. So all you have to do to mirror it is set up a cron job like:

rsync -az --delete rsync://master.pgxn.org/pgxn /path/to/pgxn

If the destination directory is under an HTTP root, you’re done. Otherwise, throw a web server over it and then you’re done.

Of course, the web server isn’t really necessary yet. Soon it will be the interface for downloading distributions and metadata from the mirrors. Right now, the mirrors just have two files in them (a README and an index.html). This is just the start of things. The next step is to figure out the directory structure for the mirrors. I’m working on that right now and will soon be asking for feedback. Watch this space for details!

]]>