Skip to content

Latest commit

 

History

History
85 lines (64 loc) · 2.55 KB

README.md

File metadata and controls

85 lines (64 loc) · 2.55 KB
Raw

Setup of Taginfo Test Server

The files in this directory are used to set up a Taginfo test server. You will need an Hetzner cloud account and the hcloud command line tool.

Setting up the server

Here are the steps needed to install the master server:

  • Go to the Hetzner Cloud console and log in.
  • Add a new project taginfo.
  • Add one or more of your ssh public keys. Name one admin.
  • Add a new API token to the project. Put the API token somewhere safe.
  • Create a hcloud context on your own machine: hcloud context create taginfo. It will ask you for the token you have just created.
  • Create a new cloud server. (You can use the --ssh-key option multiple times if you have several keys to add.) On your own machine run:
hcloud server create \
    --name taginfo \
    --location nbg1 \
    --type cx21 \
    --image debian-11 \
    --ssh-key admin

or

hcloud server create \
    --name taginfo \
    --location nbg1 \
    --type cx21 \
    --image ubuntu-22.04 \
    --ssh-key admin

(Depending an what you want to do, you might need a smaller or bigger virtual machine. For updating from a planet file you probably need a cx51 with 32 GB RAM. If you are just running the UI, a smaller machine is fine.)

  • You should now be able to log into the server as root (hcloud server ssh taginfo).
  • Copy the script init.sh to the new server and run it as root user:
IP=$(hcloud server ip taginfo)
echo $IP
scp init.sh root@$IP:/tmp/
ssh -t root@$IP /tmp/init.sh
  • If his script runs through without errors, you are done with the update of the server and you can now log in as the robot user:
hcloud server ssh -u robot taginfo

Using

There are basically two different use cases for this test server:

  • You want to test or work on the taginfo update mechanism
  • You want to test or work on the taginfo user interface

Test or work on the update mechanism

  • Run compile-tools.sh once to download and compile the C++ tools that a taginfo update needs.
  • Run download-planet.sh to download the current planet PBF file into the right place or download some other OSM file.
  • Check that the config file /srv/taginfo/taginfo-config.json is correct.
  • Run run-update.sh to do an update cycle at least once. For all stats to be updated correctly you have to run it twice.

Test or work on the user interface

  • Run download-databases.sh to download all databases from the main OSMF server to get you set up quickly.