Merge pull request #2 from shikharish/backup

.env.template

@@ -2,4 +2,13 @@ DATABASE_HOST=localhost
 DATABASE_PORT=5432
 DATABASE_NAME=databaseofbabel
 DATABASE_USERNAME=metakgp_user
-DATABASE_PASSWORD=somerandomstringfordevelopment
+DATABASE_PASSWORD=somerandomstringfordevelopment
+# Path to secret file containing database access information(username, password, etc.)
+DBPASSFILE=
+# Dropbox api variables(used for storing backups)
+DROPBOX_APP_KEY=
+DROPBOX_APP_SECRET=
+DROPBOX_ACCESS_TOKEN=
+DROPBOX_REFRESH_TOKEN=
+# Slack Webhook URL used to send incident reports and errors(like Dropbox backup failure)
+SLACK_INCIDENTS_WH_URL=

.gitignore

@@ -1 +1,2 @@
-.env
+.env
+.pgpass

Dockerfile

@@ -1 +1 @@
-FROM postgres:latest
+FROM postgres:16

backup/Dockerfile

@@ -0,0 +1,28 @@
+FROM postgres:16
+
+RUN apt-get update && apt-get install -y curl python3 python3-pip gzip
+
+ENV TZ=Asia/Kolkata
+RUN ln -snf /usr/share/zoneinfo/$TZ /etc/localtime && echo $TZ > /etc/timezone
+
+# install supercronic as cron replacement : https://github.com/aptible/supercronic?tab=readme-ov-file#why-supercronic
+ENV SUPERCRONIC_URL=https://github.com/aptible/supercronic/releases/download/v0.2.30/supercronic-linux-amd64 \
+    SUPERCRONIC=supercronic-linux-amd64 \
+    SUPERCRONIC_SHA1SUM=9f27ad28c5c57cd133325b2a66bba69ba2235799
+
+RUN curl -fsSLO "$SUPERCRONIC_URL" \
+    && echo "${SUPERCRONIC_SHA1SUM}  ${SUPERCRONIC}" | sha1sum -c - \
+    && chmod +x "$SUPERCRONIC" \
+    && mv "$SUPERCRONIC" "/usr/local/bin/${SUPERCRONIC}" \
+    && ln -s "/usr/local/bin/${SUPERCRONIC}" /usr/local/bin/supercronic
+
+WORKDIR /root
+
+ARG DBPASSFILE
+
+COPY ./ ./backup
+COPY ${DBPASSFILE} ./.pgpass
+
+RUN pip3 install -qr ./backup/requirements.txt --break-system-packages
+
+CMD ["supercronic", "/root/backup/crontab"]

backup/backup_to_dropbox.py

@@ -0,0 +1,51 @@
+#!/usr/bin/env python3
+
+import dropbox
+import traceback
+import sys
+import os
+
+file_name = sys.argv[1]
+
+app_key = os.environ["DROPBOX_APP_KEY"]
+app_secret = os.environ["DROPBOX_APP_SECRET"]
+access_token = os.environ["DROPBOX_ACCESS_TOKEN"]
+refresh_token = os.environ["DROPBOX_REFRESH_TOKEN"]
+client = dropbox.Dropbox(
+    app_key=app_key,
+    app_secret=app_secret,
+    oauth2_access_token=access_token,
+    oauth2_refresh_token=refresh_token,
+)
+
+with open(file_name, "rb") as f:
+    chunksize = 32 * 1024 * 1024
+
+    try:
+        if os.path.getsize(file_name) < chunksize:
+            result = client.files_upload(f.read(), "/" + file_name)
+            print(result)
+
+        else:
+            next_chunk = f.read(chunksize)
+
+            session = client.files_upload_session_start(next_chunk)
+            uploaded = len(next_chunk)
+
+            next_chunk = f.read(chunksize)
+            cursor = dropbox.files.UploadSessionCursor(session.session_id, uploaded)
+
+            print("Uploaded: ", float(uploaded) / (1024 * 1024), "MB")
+            while next_chunk:
+                client.files_upload_session_append_v2(next_chunk, cursor)
+                uploaded += len(next_chunk)
+                cursor = dropbox.files.UploadSessionCursor(session.session_id, uploaded)
+                print("Uploaded: ", float(uploaded) / (1024 * 1024), "MB")
+
+                next_chunk = f.read(chunksize)
+
+            commit_info = dropbox.files.CommitInfo(path="/" + file_name)
+            result = client.files_upload_session_finish(f.read(), cursor, commit_info)
+            print(result)
+    except Exception as e:
+        sys.exit(traceback.format_exc())

backup/crontab

@@ -0,0 +1 @@
+20 4 * * * /root/backup/run_backup.sh >> /var/log/backups.log 2>&1

backup/requirements.txt

@@ -0,0 +1 @@
+dropbox

backup/rotate_backups.py

@@ -0,0 +1,56 @@
+#!/usr/bin/env python3
+
+import dropbox
+import os
+import traceback
+import sys
+import datetime
+
+app_key = os.environ["DROPBOX_APP_KEY"]
+app_secret = os.environ["DROPBOX_APP_SECRET"]
+access_token = os.environ["DROPBOX_ACCESS_TOKEN"]
+refresh_token = os.environ["DROPBOX_REFRESH_TOKEN"]
+client = dropbox.Dropbox(
+    app_key=app_key,
+    app_secret=app_secret,
+    oauth2_access_token=access_token,
+    oauth2_refresh_token=refresh_token,
+)
+
+BACKUP_FOLDER_PATH = ""
+has_more_files = True
+cursor = None
+result = None
+files = []
+now = datetime.datetime.now()
+counter = 0
+
+try:
+    print("Fetching files from Dropbox...")
+    while has_more_files:
+        if cursor is None:
+            result = client.files_list_folder(BACKUP_FOLDER_PATH)
+        else:
+            result = client.files_list_folder_continue(cursor=cursor)
+        cursor = result.cursor
+        has_more_files = result.has_more
+        files.extend(result.entries)
+
+    for file in files:
+        if file.name.find("dob_dump") == -1:
+            files.remove(file)
+
+    number_of_files = len(files)
+    print(f"{number_of_files} backup files found.")
+
+    print("Starting rotation")
+    for file in files:
+        file_timestamp = file.client_modified
+        days_old = (now - file_timestamp).days
+        if days_old > 30 and (number_of_files - counter) > 30:
+            client.files_delete(file.path_display)
+            counter += 1
+
+    print(f"{counter} backup file(s) successfully deleted.")
+except Exception as e:
+    sys.exit(traceback.format_exc())

backup/run_backup.sh

@@ -0,0 +1,53 @@
+#!/usr/bin/env bash
+
+echo -e "\n-- Database of Babel Backup and Rotation START --\n"
+
+timestamp=$(date +%Y_%m_%d_%H_%M_%S)
+backup_to_dropbox="python3 /root/backup/backup_to_dropbox.py"
+rotate_backup="python3 /root/backup/rotate_backups.py"
+backups_dir="/root/backup/backups"
+dump_file="dob_dump_$timestamp.sql"
+backup_file="$dump_file.gz"
+
+mkdir -p "${backups_dir}"
+cd "$backups_dir"
+
+if ! pg_dumpall -U $POSTGRES_USER --no-password >$dump_file; then
+	echo "pgdump failure!"
+	exit 1
+fi
+
+gzip $dump_file
+
+# Dropbox backup rotation
+if ! $rotate_backup; then
+	echo "failed to rotate backups!"
+	# Notify Slack
+	if [[ -n "$SLACK_INCIDENTS_WH_URL" ]]; then
+		curl -s -H 'content-type: application/json' \
+			-d "{ \"text\": \"🔴DoB Dropbox backup rotation failure on ${timestamp}\" }" \
+			"$SLACK_INCIDENTS_WH_URL"
+	fi
+	exit 1
+fi
+
+# Delete local backups older than one week
+for file in ./*.gz; do
+	if [ $(($(date +%s) - $(date -r $file +%s))) -gt 604800 ]; then
+		rm "$file"
+	fi
+done
+
+# Backup to Dropbox
+if ! $backup_to_dropbox $backup_file; then
+	echo "Dropbox backup failure!"
+	# Notify Slack
+	if [[ -n "$SLACK_INCIDENTS_WH_URL" ]]; then
+		curl -s -H 'content-type: application/json' \
+			-d "{ \"text\": \"🔴DoB Dropbox backup failure\nCould not upload \`${backup_file}\`.\" }" \
+			"$SLACK_INCIDENTS_WH_URL"
+	fi
+	exit 1
+fi
+
+echo -e "\n-- Database of Babel Backup and Rotation END --\n"

docker-compose.yml

@@ -1,26 +1,48 @@
 services:
-    database:
-        image: metakgporg/dob
-        container_name: dob
-        build: .
-        networks:
-            metaploy-private-network:
-                aliases:
-                    - postgres-of-babel
-        environment:
-            - POSTGRES_USER=${DATABASE_USERNAME}
-            - POSTGRES_PASSWORD=${DATABASE_PASSWORD}
-            - POSTGRES_DB=${DATABASE_NAME}
-            - PGPORT=${DATABASE_PORT}
-            - PGHOST=${DATABASE_HOST}
-        volumes:
-            - db-volume:/var/lib/postgresql/data
-        restart: always
+  database:
+    image: metakgporg/dob
+    container_name: dob
+    build: .
+    networks:
+      metaploy-private-network:
+        aliases:
+          - postgres-of-babel
+    environment:
+      - POSTGRES_USER=${DATABASE_USERNAME}
+      - POSTGRES_PASSWORD=${DATABASE_PASSWORD}
+      - POSTGRES_DB=${DATABASE_NAME}
+      - PGPORT=${DATABASE_PORT}
+      - PGHOST=${DATABASE_HOST}
+    volumes:
+      - db-volume:/var/lib/postgresql/data
+    restart: always
+
+  backup:
+    build:
+      context: ./backup
+      args:
+        - DBPASSFILE=${DBPASSFILE}
+    restart: always
+    networks:
+      metaploy-private-network:
+    depends_on:
+      - database
+    environment:
+      - PGHOST=database
+      - POSTGRES_USER=${DATABASE_USERNAME}
+      - PGPASSFILE=/root/.pgpass
+      - DROPBOX_APP_KEY=${DROPBOX_APP_KEY}
+      - DROPBOX_APP_SECRET=${DROPBOX_APP_SECRET}
+      - DROPBOX_ACCESS_TOKEN=${DROPBOX_ACCESS_TOKEN}
+      - DROPBOX_REFRESH_TOKEN=${DROPBOX_REFRESH_TOKEN}
+      - SLACK_INCIDENTS_WH_URL=${SLACK_INCIDENTS_WH_URL}
+    volumes:
+      - db-volume:/var/lib/postgresql/data
 
 networks:
-    metaploy-private-network:
-        external: true
-        name: metaploy-private-network
+  metaploy-private-network:
+    external: true
+    name: metaployprivate- -network
 
 volumes:
-    db-volume:
+  db-volume: