Sign in Join
Wiki copytemporary

Mirrors of this wiki

Anyone can keep a copy of this wiki on their own web server. Mirrors keep the pages readable when this site is busy or down, and they take load off it. A mirror is plain files: no database, no program to run on each visit, nothing to sign in to.

Mirrors you can read now

Run a mirror

You need a Linux or BSD machine with a web server, about 500 MB of disk, and bash, curl and cron.

  1. Download the script:
    curl -O https://pascal.chat/static/pascal-mirror.sh
  2. Read it, then run it with the folder your web server serves and that folder's address on the web:
    bash pascal-mirror.sh /var/www/html/pascal-wiki https://example.org/pascal-wiki/
  3. That is all. The script copies the wiki (about 400 MB the first time), checks every file, and sets up a cron job that brings the copy up to date every hour. This site then checks your mirror and, once an admin has said yes, lists it here.

Shared hosting or a server of your own

Shared hosting stops jobs that run long, so by default the first copy comes in parts: a run every ten minutes, whole in about an hour. There is nothing to do but wait. On a server of your own, add --fast and the whole copy comes in one run, in minutes:

bash pascal-mirror.sh --fast /var/www/html/pascal-wiki https://example.org/pascal-wiki/

Either way, once the copy is whole, cron brings it up to date every hour and fetches only what changed.

Leave out the address and the copy still works; it is only not listed. To stop, run bash pascal-mirror.sh remove /var/www/html/pascal-wiki: it takes away the cron job and the listing, and leaves the files for you to delete.

Every link in the copy is relative, so it works at the top of a site, in a folder of one, or opened from disk. Your server never sees a password: editing and signing in happen here.

How a mirror is checked: every hour the script answers a small question from this site about a few of its files, and once a day this site reads a dozen pages from the mirror, picked at random, and compares them with its own.