reintroduce cutoffs when searching for very frequent words

[nominatim.git] / vagrant / Install-on-Ubuntu-20.sh

diff --git a/vagrant/Install-on-Ubuntu-20.sh b/vagrant/Install-on-Ubuntu-20.sh
index 292947d179ec91ecd12a265e852c98be5711a215..57361ec7755cf1c2047120d6c3d1d768b680e1b6 100755 (executable)
--- a/vagrant/Install-on-Ubuntu-20.sh
+++ b/vagrant/Install-on-Ubuntu-20.sh
@@ -1,8 +1,7 @@
-#!/bin/bash
+#!/bin/bash -e

 #
 # hacks for broken vagrant box      #DOCS:
 sudo rm -f /var/lib/dpkg/lock       #DOCS:
 #
 # hacks for broken vagrant box      #DOCS:
 sudo rm -f /var/lib/dpkg/lock       #DOCS:

-sudo update-locale LANG=en_US.UTF-8 #DOCS:

 export APT_LISTCHANGES_FRONTEND=none #DOCS:
 export DEBIAN_FRONTEND=noninteractive #DOCS:
 
 export APT_LISTCHANGES_FRONTEND=none #DOCS:
 export DEBIAN_FRONTEND=noninteractive #DOCS:
 


@@ -14,27 +13,27 @@ export DEBIAN_FRONTEND=noninteractive #DOCS:
 #
 # These instructions expect that you have a freshly installed Ubuntu 20.04.
 #
 #
 # These instructions expect that you have a freshly installed Ubuntu 20.04.
 #

-# Make sure all packages are are up-to-date by running:
+# Make sure all packages are up-to-date by running:

 #
 
 #
 

-#DOCS:    :::sh
-    sudo apt-get \ #DOCS:
-        -o DPkg::options::="--force-confdef" -o DPkg::options::="--force-confold" \ #DOCS:
-        --allow-downgrades --allow-remove-essential --allow-change-held-packages \ #DOCS:
-        -fuy install grub-pc #DOCS:
-    sudo apt update -qq
+    sudo apt-get update -qq

 
 # Now you can install all packages needed for Nominatim:
 
 
 # Now you can install all packages needed for Nominatim:
 

-    sudo apt install -y php-cgi
-    sudo apt install -y build-essential cmake g++ libboost-dev libboost-system-dev \
+    sudo apt-get install -y build-essential cmake g++ libboost-dev libboost-system-dev \

                         libboost-filesystem-dev libexpat1-dev zlib1g-dev \
                         libboost-filesystem-dev libexpat1-dev zlib1g-dev \

-                        libbz2-dev libpq-dev libproj-dev \
-                        postgresql-server-dev-12 postgresql-12-postgis-3 \
+                        libbz2-dev libpq-dev liblua5.3-dev lua5.3 lua-dkjson \
+                        nlohmann-json3-dev postgresql-12-postgis-3 \

                         postgresql-contrib-12 postgresql-12-postgis-3-scripts \
                         postgresql-contrib-12 postgresql-12-postgis-3-scripts \

-                        php php-pgsql php-intl libicu-dev python3-dotenv \
-                        python3-psycopg2 python3-psutil python3-jinja2 python3-icu git \
-                        python3-argparse-manpage
+                        libicu-dev python3-dotenv \
+                        python3-psycopg2 python3-psutil python3-jinja2 python3-pip \
+                        python3-icu python3-datrie python3-yaml git
+
+# Some of the Python packages that come with Ubuntu 20.04 are too old, so
+# install the latest version from pip:
+
+    pip3 install --user sqlalchemy asyncpg
+

 
 #
 # System Configuration
 
 #
 # System Configuration


@@ -56,11 +55,18 @@ export DEBIAN_FRONTEND=noninteractive #DOCS:
 #
 # You may find a more suitable location if you wish.
 #
 #
 # You may find a more suitable location if you wish.
 #

+# The following instructions assume you are logged in as this user.
+# You can also switch to the user with:
+#
+#     sudo -u nominatim bash
+#

 # To be able to copy and paste instructions from this manual, export
 # user name and home directory now like this:
 #
 # To be able to copy and paste instructions from this manual, export
 # user name and home directory now like this:
 #

+if [ "x$USERNAME" == "x" ]; then #DOCS:

     export USERNAME=vagrant        #DOCS:    export USERNAME=nominatim
     export USERHOME=/home/vagrant  #DOCS:    export USERHOME=/srv/nominatim
     export USERNAME=vagrant        #DOCS:    export USERNAME=nominatim
     export USERHOME=/home/vagrant  #DOCS:    export USERHOME=/srv/nominatim

+fi                                 #DOCS:

 #
 # **Never, ever run the installation as a root user.** You have been warned.
 #
 #
 # **Never, ever run the installation as a root user.** You have been warned.
 #


@@ -72,14 +78,17 @@ export DEBIAN_FRONTEND=noninteractive #DOCS:
 # ---------------------
 #
 # Tune the postgresql configuration, which is located in 
 # ---------------------
 #
 # Tune the postgresql configuration, which is located in 

-# `/etc/postgresql/12/main/postgresql.conf`. See section *Postgres Tuning* in
-# [the installation page](../admin/Installation.md#postgresql-tuning)
+# `/etc/postgresql/12/main/postgresql.conf`. See section *Tuning the PostgreSQL database*
+# in [the installation page](../admin/Installation.md#tuning-the-postgresql-database)

 # for the parameters to change.
 #
 # Restart the postgresql service after updating this config file.
 
 # for the parameters to change.
 #
 # Restart the postgresql service after updating this config file.
 

+if [ "x$NOSYSTEMD" == "xyes" ]; then  #DOCS:
+    sudo pg_ctlcluster 12 main start  #DOCS:
+else                                  #DOCS:

     sudo systemctl restart postgresql
     sudo systemctl restart postgresql

-
+fi                                    #DOCS:

 #
 # Finally, we need to add two postgres users: one for the user that does
 # the import and another for the webserver which should access the database
 #
 # Finally, we need to add two postgres users: one for the user that does
 # the import and another for the webserver which should access the database


@@ -100,7 +109,7 @@ export DEBIAN_FRONTEND=noninteractive #DOCS:
 #
 if [ "x$1" == "xyes" ]; then  #DOCS:    :::sh
     cd $USERHOME
 #
 if [ "x$1" == "xyes" ]; then  #DOCS:    :::sh
     cd $USERHOME

-    git clone --recursive git://github.com/openstreetmap/Nominatim.git
+    git clone --recursive https://github.com/openstreetmap/Nominatim.git

     cd Nominatim
 else                               #DOCS:
     cd $USERHOME/Nominatim         #DOCS:
     cd Nominatim
 else                               #DOCS:
     cd $USERHOME/Nominatim         #DOCS:


@@ -110,7 +119,7 @@ fi                                 #DOCS:
 # download the country grid:
 
 if [ ! -f data/country_osm_grid.sql.gz ]; then       #DOCS:    :::sh
 # download the country grid:
 
 if [ ! -f data/country_osm_grid.sql.gz ]; then       #DOCS:    :::sh

-    wget -O data/country_osm_grid.sql.gz https://www.nominatim.org/data/country_grid.sql.gz
+    wget -O data/country_osm_grid.sql.gz https://nominatim.org/data/country_grid.sql.gz

 fi                                 #DOCS:
 
 # The code must be built in a separate directory. Create this directory,
 fi                                 #DOCS:
 
 # The code must be built in a separate directory. Create this directory,


@@ -124,43 +133,107 @@ fi                                 #DOCS:
 
 # Nominatim is now ready to use. You can continue with
 # [importing a database from OSM data](../admin/Import.md). If you want to set up
 
 # Nominatim is now ready to use. You can continue with
 # [importing a database from OSM data](../admin/Import.md). If you want to set up

-# a webserver first, continue reading.
+# the API frontend first, continue reading.
+#
+# Setting up the Python frontend
+# ==============================

 #
 #

+# Some of the Python packages in Ubuntu are too old. Therefore run the
+# frontend from a Python virtualenv with current packages.
+#
+# To set up the virtualenv, run:
+
+#DOCS:```sh
+sudo apt-get install -y virtualenv
+virtualenv $USERHOME/nominatim-venv
+$USERHOME/nominatim-venv/bin/pip install SQLAlchemy PyICU psycopg[binary] \
+              psycopg2-binary python-dotenv PyYAML falcon uvicorn gunicorn
+#DOCS:```
+
+# Next you need to create a systemd job that runs Nominatim on gunicorn.
+# First create a systemd job that manages the socket file:
+
+#DOCS:```sh
+sudo tee /etc/systemd/system/nominatim.socket << EOFSOCKETSYSTEMD
+[Unit]
+Description=Gunicorn socket for Nominatim
+
+[Socket]
+ListenStream=/run/nominatim.sock
+SocketUser=www-data
+
+[Install]
+WantedBy=multi-user.target
+EOFSOCKETSYSTEMD
+#DOCS:```
+
+# Then create the service for Nominatim itself.
+
+#DOCS:```sh
+sudo tee /etc/systemd/system/nominatim.service << EOFNOMINATIMSYSTEMD
+[Unit]
+Description=Nominatim running as a gunicorn application
+After=network.target
+Requires=nominatim.socket
+
+[Service]
+Type=simple
+Environment="PYTHONPATH=/usr/local/lib/nominatim/lib-python/"
+User=www-data
+Group=www-data
+WorkingDirectory=$USERHOME/nominatim-project
+ExecStart=$USERHOME/nominatim-venv/bin/gunicorn -b unix:/run/nominatim.sock -w 4 -k uvicorn.workers.UvicornWorker nominatim.server.falcon.server:run_wsgi
+ExecReload=/bin/kill -s HUP \$MAINPID
+StandardOutput=append:/var/log/gunicorn-nominatim.log
+StandardError=inherit
+PrivateTmp=true
+TimeoutStopSec=5
+KillMode=mixed
+
+[Install]
+WantedBy=multi-user.target
+EOFNOMINATIMSYSTEMD
+#DOCS:```
+
+# Activate the services:
+
+if [ "x$NOSYSTEMD" != "xyes" ]; then  #DOCS:
+    sudo systemctl daemon-reload
+    sudo systemctl enable nominatim.socket
+    sudo systemctl start nominatim.socket
+    sudo systemctl enable nominatim.service
+fi                                    #DOCS:
+
+

 # Setting up a webserver
 # ======================
 #
 # Setting up a webserver
 # ======================
 #

-# The webserver should serve the php scripts from the website directory of your
-# [project directory](../admin/import.md#creating-the-project-directory).
-# Therefore set up a project directory and populate the website directory:
+# The webserver is only needed as a proxy between the public interface
+# and the gunicorn service.
+#
+# The frontend will need configuration information from the project
+# directory, which will be populated later
+# [during the import process](../admin/Import.md#creating-the-project-directory)
+# Already create the project directory itself now:
+

 
     mkdir $USERHOME/nominatim-project
 
     mkdir $USERHOME/nominatim-project

-    cd $USERHOME/nominatim-project
-    nominatim refresh --website

 
 

-#
+

 # Option 1: Using Apache
 # ----------------------
 #
 if [ "x$2" == "xinstall-apache" ]; then #DOCS:
 # Option 1: Using Apache
 # ----------------------
 #
 if [ "x$2" == "xinstall-apache" ]; then #DOCS:

-#
-# Apache has a PHP module that can be used to serve Nominatim. To install them
-# run:
+# First install apache itself and enable the proxy module:

 
 

-    sudo apt install -y apache2 libapache2-mod-php
+    sudo apt-get install -y apache2
+    sudo a2enmod proxy_http

 
 

-# You need to create an alias to the website directory in your apache
-# configuration. Add a separate nominatim configuration to your webserver:
+# To set up proxying for Apache add the following configuration:

 
 #DOCS:```sh
 sudo tee /etc/apache2/conf-available/nominatim.conf << EOFAPACHECONF
 
 #DOCS:```sh
 sudo tee /etc/apache2/conf-available/nominatim.conf << EOFAPACHECONF

-<Directory "$USERHOME/nominatim-project/website">
-  Options FollowSymLinks MultiViews
-  AddType text/html   .php
-  DirectoryIndex search.php
-  Require all granted
-</Directory>
-
-Alias /nominatim $USERHOME/nominatim-project/website
+ProxyPass /nominatim "unix:/run/nominatim.sock|http://localhost/"

 EOFAPACHECONF
 #DOCS:```
 
 EOFAPACHECONF
 #DOCS:```
 


@@ -169,7 +242,11 @@ EOFAPACHECONF
 #
 
     sudo a2enconf nominatim
 #
 
     sudo a2enconf nominatim

+if [ "x$NOSYSTEMD" == "xyes" ]; then  #DOCS:
+    sudo apache2ctl start             #DOCS:
+else                                  #DOCS:

     sudo systemctl restart apache2
     sudo systemctl restart apache2

+fi                                    #DOCS:

 
 # The Nominatim API is now available at `http://localhost/nominatim/`.
 
 
 # The Nominatim API is now available at `http://localhost/nominatim/`.
 


@@ -181,33 +258,9 @@ fi   #DOCS:
 #
 if [ "x$2" == "xinstall-nginx" ]; then #DOCS:
 
 #
 if [ "x$2" == "xinstall-nginx" ]; then #DOCS:
 

-# Nginx has no native support for php scripts. You need to set up php-fpm for
-# this purpose. First install nginx and php-fpm:
-
-    sudo apt install -y nginx php-fpm
-
-# You need to configure php-fpm to listen on a Unix socket.
+# First install nginx itself:

 
 

-#DOCS:```sh
-sudo tee /etc/php/7.4/fpm/pool.d/www.conf << EOF_PHP_FPM_CONF
-[www]
-; Replace the tcp listener and add the unix socket
-listen = /var/run/php7.4-fpm.sock
-
-; Ensure that the daemon runs as the correct user
-listen.owner = www-data
-listen.group = www-data
-listen.mode = 0666
-
-; Unix user of FPM processes
-user = www-data
-group = www-data
-
-; Choose process manager type (static, dynamic, ondemand)
-pm = ondemand
-pm.max_children = 5
-EOF_PHP_FPM_CONF
-#DOCS:```
+    sudo apt-get install -y nginx

 
 # Then create a Nginx configuration to forward http requests to that socket.
 
 
 # Then create a Nginx configuration to forward http requests to that socket.
 


@@ -218,38 +271,27 @@ server {
     listen [::]:80 default_server;
 
     root $USERHOME/nominatim-project/website;
     listen [::]:80 default_server;
 
     root $USERHOME/nominatim-project/website;

-    index search.php index.html;
-    location / {
-        try_files \$uri \$uri/ @php;
-    }
-
-    location @php {
-        fastcgi_param SCRIPT_FILENAME "\$document_root\$uri.php";
-        fastcgi_param PATH_TRANSLATED "\$document_root\$uri.php";
-        fastcgi_param QUERY_STRING    \$args;
-        fastcgi_pass unix:/var/run/php7.4-fpm.sock;
-        fastcgi_index index.php;
-        include fastcgi_params;
-    }
-
-    location ~ [^/]\.php(/|$) {
-        fastcgi_split_path_info ^(.+?\.php)(/.*)$;
-        if (!-f \$document_root\$fastcgi_script_name) {
-            return 404;
-        }
-        fastcgi_pass unix:/var/run/php7.4-fpm.sock;
-        fastcgi_index search.php;
-        include fastcgi.conf;
+    index /search;
+
+    location /nominatim/ {
+            proxy_set_header Host \$http_host;
+            proxy_set_header X-Forwarded-For \$proxy_add_x_forwarded_for;
+            proxy_set_header X-Forwarded-Proto \$scheme;
+            proxy_redirect off;
+            proxy_pass http://unix:/run/nominatim.sock:/;

     }
 }
 EOF_NGINX_CONF
 #DOCS:```
 
     }
 }
 EOF_NGINX_CONF
 #DOCS:```
 

-#

 # Enable the configuration and restart Nginx
 #
 
 # Enable the configuration and restart Nginx
 #
 

-    sudo systemctl restart php7.4-fpm nginx
+if [ "x$NOSYSTEMD" == "xyes" ]; then  #DOCS:
+    sudo /usr/sbin/nginx &            #DOCS:
+else                                  #DOCS:
+    sudo systemctl restart nginx
+fi                                    #DOCS:

 
 # The Nominatim API is now available at `http://localhost/`.
 
 
 # The Nominatim API is now available at `http://localhost/`.