This Markdown page is for the GPlates Web Service developers only. It contains some technical notes which may seem confusing for other people. Contact the development team if you would like to know some details about anything in this page.
docker pull redis
docker run --name gplates-redis --network gplates-net -d redis
docker run -it --network gplates-net --rm redis redis-cli -h gplates-redis
- ./docker/start-postgis.sh
- ./docker/start-gws-dev.sh
- ./test/test-dev-server.sh
- ./test/test-pro-server.sh
- http://localhost:18000/
You need to specify a writable folder for the log files in settings.py. By default, the folder is /gws/log.
sudo a2enmod proxy_balancer
sudo a2enmod lbmethod_bytraffic
sudo a2enmod lbmethod_byrequests
sudo systemctl restart apache2
<proxy balancer://serverpool>
BalancerMember http://localhost:18000
BalancerMember https://gws-nci.gplates.org
ProxySet lbmethod=bytraffic
</proxy>
ProxyPass / "balancer://serverpool/"
ProxyPassReverse / "balancer://serverpool/"
https://gws.gplates.org/lb-manager
docker compose run --rm -d --service-ports gplates-postgis
docker run -d -v `pwd`:/gws -p 18000:80 --network gplates-net --restart always gplates/gws
You can change the numbers at docker/gws.conf. Each wsgi process may require 1.5~2.0GB memory. Check how much memory your computer has.
- install Apache2
- install certbot
- add gws.conf and gws-ssl.conf on host computer and redirect https traffic to http://localhost:18000
- sudo a2enmod proxy && sudo a2enmod proxy_http && sudo a2enmod ssl
- sudo systemctl restart apache2
- sudo certbot --apache
- go into gws docker container and change number of processes
- command + shif + space and click
magick -delay 20 $(ls *.png | sort -Vr) out.gif
magick out.gif -coalesce temporary.gif
(make .gif smaller - step 1)magick temporary.gif -resize 504x315 gws-location-reconstruction.gif
(make .gif smaller - step 2)