FoolFuuka/Start

From Bibliotheca Anonoma

We should start using systemd instead of screen, but here is the procedure in case you have a system without systemd (such as Ubuntu 14.04 LTS)

Important Directories[edit]

  • /home/asagi - location of asagi and related files
  • /var/www/foolfuuka - location of desuarchive foolfuuka php files
  • /var/www/rbt-foolfuuka - location of rbt.asia foolfuuka php files
  • /home/atc/ - location of your own user's php scripts for deleting content. We use atc as the example user.
  • /var/www/foolfuuka/app/foolz/foolframe/logs - location of log files


Boot Procedure[edit]

/etc/rc.local is ran, has some configurations for tokudb. It disables hugepages just in case a kernel update breaks it.

  • Startup script is /home/asagi/start.sh
    • Starts asagi (asagi)
    • Starts intel sharing & ban fetching for Desu (intel / bans)
    • Starts intel sharing & ban fetching for RBT (rintel / rbans)
    • Stats is disabled, but you can uncomment those (stats / rstats)

Restarting Asagi by screen sessions[edit]

The commands to start each service of FoolFuuka/Asagi are ordered by screen commands in the following file:

/home/asagi/start.sh

Find the screen session of the service you want to restart with htop or something. Then kill the process ID of it.

Redis Cache[edit]

Caches results of CPU intensive calculations. They do expire. Generally shouldn’t have a problem but if it does, restart.

   sudo systemctl restart redis


When to stop the Asagi dumper[edit]

Stop the Asagi dumper whenever you’re importing an external dump file (Anything heavy you do with the database)

Exception: Backups, they don’t lock the db

Need to grab past week's posts from 4chan.org Archives.json[edit]

Sometimes your archiver will not be running and you want to fetch the past week or so's worth of threads and images. Use the asagi.arch instance to do so.

   sudo su asagi
   cd /home/asagi/asagi.arch/
   java -jar asagi.jar
  • let it run for an hour or two, then Ctrl + C it (based on the original dumper, so it will keep refreshing archives.json file)
  • anything made in the last 7 days will be grabbed from this