Инструкция по установке DreamDocs

export DD_VERSION="<version>"
export DREAMDOCS_RELEASE="/srv/dreamdocs-${DD_VERSION}"
export DREAMDOCS_HOME="/srv/dreamdocs"
export SERVER_IP="$(hostname -I | awk '{print $1}')"
export SERVER_NAME="<server_name>"

sudo apt update -y
sudo apt autoremove -y
sudo apt clean

sudo groupadd --system dream || true
sudo useradd --system --create-home --shell /bin/bash --gid dream dream || true
sudo usermod -aG sudo dream

sudo mkdir -p "$DREAMDOCS_RELEASE" /srv/downloads /media/ocr
sudo ln -sfn "$DREAMDOCS_RELEASE" "$DREAMDOCS_HOME"
sudo chown -R dream:dream "$DREAMDOCS_RELEASE" "$DREAMDOCS_HOME" /srv/downloads /media/ocr
sudo chmod -R u+rwX,g+rwX,o-rwx "$DREAMDOCS_RELEASE" /srv/downloads /media/ocr

sudo su - dream

sudo apt install -y zlib1g-dev libncurses5-dev libgdbm-dev libnss3-dev   libssl-dev libsqlite3-dev libreadline-dev libffi-dev curl libbz2-dev
sudo apt install -y libncursesw5-dev tk-dev libc6-dev libvlccore-dev   libvlc-dev liblzma-dev lzma cmake
sudo apt install -y --reinstall pkg-config cmake-data
sudo apt install -y libpoppler-cpp-dev
sudo apt install -y --no-install-recommends libreoffice-calc libreoffice-common   libzbar0 poppler-utils screen
sudo apt install -y tesseract-ocr tesseract-ocr-rus zbar-tools git unzip

cd /srv/downloads
curl -O <system_packages_archive_url>
tar -xvzf <system_packages_archive_name>.tgz
cd <system_packages_directory>
# В install.sh должен быть указан корректный путь до каталога packages.
./install.sh

python3 --version || true
nginx -v || true
tesseract --version || true
pdftotext -v || true
soffice --version || true

cd "$DREAMDOCS_HOME"
mkdir -p data/db data/files data/lock data/logs data/models data/search data/dump_workspace

cd "$DREAMDOCS_HOME"
curl -O <dreamdocs_distribution_url>/dreamdocs-<version>-full.zip

cd "$DREAMDOCS_HOME"
unzip dreamdocs-<version>-full.zip -d "$DREAMDOCS_HOME"
rm dreamdocs-<version>-full.zip

ln -rsf "$DREAMDOCS_HOME/docrender/docrender" "$DREAMDOCS_HOME/dreamocr"
ln -rsf "$DREAMDOCS_HOME/docrender/docrender" "$DREAMDOCS_HOME/docaml/backend"
ln -rsf "$DREAMDOCS_HOME/docrender/docrender" "$DREAMDOCS_HOME/doca/backend"
ln -rsf "$DREAMDOCS_HOME/dreamocr/dreamocr" "$DREAMDOCS_HOME/docaml/backend"
ln -rsf "$DREAMDOCS_HOME/docanorm" "$DREAMDOCS_HOME/doca/backend"
ln -rsf "$DREAMDOCS_HOME/docanorm" "$DREAMDOCS_HOME/docaml/backend"

ln -rs "$DREAMDOCS_HOME/dreamocr/dreamocr" "$DREAMDOCS_HOME/docaml/backend"

cd "$DREAMDOCS_HOME"
cp -n doca.yaml doca.yaml.backup 2>/dev/null || true
mcedit doca.yaml

OCR_KEY: "<ocr_token>"
DREAMOCR_DATA_PATH: "/srv/dreamdocs/dreamocr.data"
ALLOWED_HOSTS:
  - "localhost"
  - "127.0.0.1"
  - "0.0.0.0"
  - "<server_ip>"
  - "<server_name>"
LICENSE_PATH: "<license_path>"

cd "$DREAMDOCS_HOME"
cp -n dreamocr.yaml dreamocr.yaml.backup 2>/dev/null || true
mcedit dreamocr.yaml

OCR_TIMEOUT: -1
DREAMOCR_DATA_PATH: "/srv/dreamdocs/dreamocr.data"
ALLOWED_TOKENS:
  "<ocr_token>": "<ocr_token>"
ALLOWED_HOSTS:
  - "localhost"
  - "127.0.0.1"
  - "0.0.0.0"
  - "<server_ip>"
  - "<server_name>"
LICENSE_PATH: "<license_path>"

cd /srv
mkdir -p .Python3.8
cd .Python3.8
curl -O https://www.python.org/ftp/python/3.8.10/Python-3.8.10.tar.xz
tar -xf Python-3.8.10.tar.xz
rm Python-3.8.10.tar.xz

cd /srv/.Python3.8/Python-3.8.10
./configure --enable-optimizations
make -j 4
sudo make altinstall
python3.8 -V

cd "$DREAMDOCS_HOME"
python3.8 -m venv .venv
source .venv/bin/activate
python -m pip install --upgrade pip setuptools wheel

cd "$DREAMDOCS_HOME"
pip install -r doca/requirements.txt
pip install -r docaml/requirements.txt
pip install -r dreamocr/requirements.txt

cd /srv/downloads
unzip pypack-<version>.zip -d /srv/downloads/pypack-<version>
cd "$DREAMDOCS_HOME"
pip install --no-index --find-links /srv/downloads/pypack-<version> -r doca/requirements.txt
pip install --no-index --find-links /srv/downloads/pypack-<version> -r docaml/requirements.txt
pip install --no-index --find-links /srv/downloads/pypack-<version> -r dreamocr/requirements.txt

ln -rsf "$DREAMDOCS_HOME/.venv" "$DREAMDOCS_HOME/doca"
ln -rsf "$DREAMDOCS_HOME/.venv" "$DREAMDOCS_HOME/docaml"
ln -rsf "$DREAMDOCS_HOME/.venv" "$DREAMDOCS_HOME/dreamocr"

sudo apt install -y nginx
sudo usermod -aG dream www-data || true
sudo cp /etc/nginx/nginx.conf /etc/nginx/nginx.conf.backup.$(date +%Y%m%d%H%M%S)

events {
    worker_connections 1024;
}

http {
    include /etc/nginx/mime.types;
    server {
        listen 80;
        server_name 127.0.0.1 <server_ip> <server_name>;
        root /srv/dreamdocs/doca/public/;
        index index.html;
        charset utf-8;
        source_charset utf-8;
        client_max_body_size 2500M;

        location / {
            proxy_read_timeout 600s;
            proxy_set_header Host $host;
            proxy_set_header X-Real-IP $remote_addr;
            proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
            proxy_set_header X-Forwarded-Proto http;
            proxy_cache_bypass $http_upgrade;
            proxy_pass http://127.0.0.1:8000;
            proxy_redirect off;
        }

        location ~ ^/(lib|css|js|img|fonts|v2design|favicon.ico) {
            try_files $uri /empty.html;
        }

        location /static/ {
            alias /srv/dreamdocs/doca/static/;
            try_files $uri =404;
        }

        location /files/ {
            alias /srv/dreamdocs/data/files/;
            try_files $uri =404;
        }
    }
}

events {
    worker_connections 1024;
}

http {
    include /etc/nginx/mime.types;
    server {
        listen 443 ssl;
        server_name <server_name>;

        ssl_certificate /path/to/certs/app.crt;
        ssl_certificate_key /path/to/certs/private/app.key;
        ssl_client_certificate /path/to/certs/rootCA.pem;
        ssl_verify_client on;
        ssl_protocols TLSv1.2 TLSv1.3;
        ssl_prefer_server_ciphers on;
        ssl_ciphers HIGH:!aNULL:!MD5;

        root /srv/dreamdocs/doca/public/;
        index index.html;
        charset utf-8;
        source_charset utf-8;
        client_max_body_size 2500M;

        location / {
            proxy_read_timeout 600s;
            proxy_set_header Host $host;
            proxy_set_header X-Real-IP $remote_addr;
            proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
            proxy_set_header X-Forwarded-Proto https;
            proxy_cache_bypass $http_upgrade;
            proxy_pass http://127.0.0.1:8000;
            proxy_redirect off;
        }

        location /static/ {
            alias /srv/dreamdocs/doca/static/;
            try_files $uri =404;
        }

        location /files/ {
            alias /srv/dreamdocs/data/files/;
            try_files $uri =404;
        }
    }
}

sudo nginx -t
sudo systemctl restart nginx
sudo systemctl status nginx --no-pager

sudo apt install -y postgresql
sudo systemctl enable postgresql
sudo systemctl status postgresql --no-pager

sudo su - postgres
psql
CREATE DATABASE <db_name>;
CREATE USER <db_user> WITH PASSWORD '<db_password>';
ALTER DATABASE <db_name> OWNER TO <db_user>;
\q
exit

sudo mcedit /etc/postgresql/<postgres_version>/main/postgresql.conf
# listen_addresses = 'localhost'  # локальный режим
# listen_addresses = '*'          # только если требуется удаленный доступ

sudo mcedit /etc/postgresql/<postgres_version>/main/pg_hba.conf
sudo systemctl restart postgresql

DATABASES:
  default:
    ENGINE: "django.db.backends.postgresql"
    NAME: "<db_name>"
    USER: "<db_user>"
    PASSWORD: "<db_password>"
    HOST: "<db_host>"
    PORT: "5432"

DATABASES:
  default:
    ENGINE: "django.db.backends.postgresql"
    NAME: "<db_name>"
    USER: "<db_user>"
    PASSWORD: ""
    HOST: "<db_host>"
    PORT: "5432"
    OPTIONS:
      sslmode: "verify-full"
      sslrootcert: "/app/certs/rootCA.pem"
      sslcert: "/app/certs/app.crt"
      sslkey: "/app/certs/app.key"

cd /srv/dreamdocs/dreamocr/backend/ocr
mcedit models.py
# В модели File: изменить max_length=32 на max_length=64 для поля hash.
# В модели Task: изменить max_length=32 на max_length=64 для поля thash.

source /srv/dreamdocs/.venv/bin/activate
cd /srv/dreamdocs/dreamocr
make migrate
cd /srv/dreamdocs/doca
make migrate

cd /srv/dreamdocs
curl -O <license_url>
# или скопируйте файл лицензии вручную:
# scp demo.license dream@<server_ip>:/srv/dreamdocs/demo.license
chmod 640 /srv/dreamdocs/*.license 2>/dev/null || true

cd /srv/dreamdocs/docaml
make gpt-server-download

LOCALGPT_ENABLE: true
DEFAULT_GPT: "localgpt"
LOCALGPT_API_KEY: "<local_gpt_api_key>"
LOCALGPT_MODEL: "mistral-7b-instruct-v0.2"

cd /srv/dreamdocs/docaml
make reload
make gpt-server-cpu
make chat

source /srv/dreamdocs/.venv/bin/activate
cd /srv/dreamdocs/dreamocr
make _build_pyx
make run

cd /srv/dreamdocs/dreamocr
make worker
make worker
make worker

source /srv/dreamdocs/.venv/bin/activate
cd /srv/dreamdocs/docaml
make run

source /srv/dreamdocs/.venv/bin/activate
cd /srv/dreamdocs/doca
make run

location /api/evs/ {
    proxy_pass http://127.0.0.1:9347/api/evs/;
    proxy_set_header Host $host;
    proxy_set_header X-Real-IP $remote_addr;
    proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
    proxy_set_header X-Forwarded-Proto https;
    proxy_set_header X-Forwarded-Referrer $http_referer;
    proxy_set_header Referer $http_referer;
    proxy_cache_bypass $http_upgrade;
    proxy_redirect off;
}

sudo nginx -t
sudo systemctl restart nginx

cd /srv/dreamdocs/doca
make broker

cd /srv/dreamdocs/docaml
make broker-worker

cd /srv/dreamdocs/data/dump_workspace
curl -O <workspace_dump_url>

cd /srv/dreamdocs/doca/backend
./manage.py load_workspace -wid=<workspace_id> -P <password> /srv/dreamdocs/data/dump_workspace

cd /srv/dreamdocs/dreamdocs
make deploy-scripts
sudo make deploy-services
sudo systemctl daemon-reload
sudo systemctl enable dreamdocs
sudo systemctl start dreamdocs
sudo systemctl status dreamdocs --no-pager

screen -ls

screen -x dreamocr-worker
screen -x doca-statemachine
# Для выхода без остановки: Ctrl+A, Ctrl+D