add logging improvement

Author: NathanEne

backup_and_restore_database/backup_database/.env.example

@@ -7,5 +7,4 @@ DB_PASSWORD=
 AWS_ENDPOINT_URL=
 AWS_ACCESS_KEY_ID=
 AWS_SECRET_ACCESS_KEY=
-AWS_DEFAULT_REGION=ca-central-1
 BUCKET_NAME=

backup_and_restore_database/backup_database/backup_and_upload_to_s3.py

@@ -1,7 +1,8 @@
 import os
 import subprocess
 import sys
-from datetime import date, datetime
+from datetime import datetime, date
+import threading
 
 DB_HOST = os.environ["DB_HOST"]
 DB_PORT = os.environ["DB_PORT"]
@@ -11,6 +12,12 @@
 ENDPOINT_URL = os.environ["AWS_ENDPOINT_URL"]
 S3_PATH = f"s3://{os.getenv('BUCKET_NAME')}/{date.today()}-{DB_NAME}.dump"
 os.environ["PGPASSWORD"] = DB_PASSWORD
+
+def log_stream(stream, prefix):
+    """Stream stderr output with timestamps"""
+    for line in stream:
+        print(f"{datetime.now()} [{prefix}] {line.decode().strip()}")
+
 pg_dump_cmd = [
     "pg_dump",
     "-h", DB_HOST,
@@ -28,22 +35,36 @@
     "--expected-size", "60000000000"
 ]
 
-print("Streaming dump into S3 bucket...")
+print(f"Starting backup to {S3_PATH} at {datetime.now()}")
+
+
+
+try:
+    dump_proc = subprocess.Popen(pg_dump_cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+
+    aws_proc = subprocess.Popen(aws_cp_cmd, stdin=dump_proc.stdout, stderr=subprocess.PIPE)
+
+    dump_proc.stdout.close()
 
-dump_proc = subprocess.Popen(pg_dump_cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+    dump_thread = threading.Thread(target=log_stream, args=(dump_proc.stderr, "pg_dump"))
+    aws_thread = threading.Thread(target=log_stream, args=(aws_proc.stderr, "aws"))
 
-aws_proc = subprocess.Popen(aws_cp_cmd, stdin=dump_proc.stdout, stderr=subprocess.PIPE)
+    dump_thread.start()
+    aws_thread.start()
 
-dump_proc.stdout.close()
+    dump_thread.join()
+    aws_thread.join()
 
-for line in dump_proc.stderr:
-    print(f"{datetime.now()} {line.decode().strip()}")
+    if dump_proc.wait() != 0:
+        print("pg_dump failed!")
+        sys.exit(1)
+    if aws_proc.wait() != 0:
+        print("AWS upload failed!")
+        sys.exit(1)
 
-# Wait for both to finish and get final stderr from aws
-_, aws_stderr = aws_proc.communicate()
+    print(f"Backup completed successfully at {datetime.now()}")
+    print(f"Saved to: {S3_PATH}")
 
-if aws_proc.returncode != 0:
-    print("[aws s3 cp] ERROR:")
-    print(aws_stderr.decode())
-if dump_proc.wait() != 0:
-    print("pg_dump exited with a non-zero status.")
+except Exception as e:
+    print(f"Backup failed at {datetime.now()}: {str(e)}")
+    sys.exit(1)

backup_and_restore_database/restore_database/.env.example

@@ -0,0 +1,11 @@
+DB_HOST=
+DB_PORT=
+DB_NAME=
+DB_USER=
+DB_PASSWORD=
+
+AWS_ENDPOINT_URL=
+AWS_ACCESS_KEY_ID=
+AWS_SECRET_ACCESS_KEY=
+BUCKET_NAME=
+BACKUP_FILE=

backup_and_restore_database/restore_database/Dockerfile

@@ -0,0 +1,34 @@
+FROM postgres:17-bookworm as pgclient
+
+FROM python:3.12-slim
+
+COPY --from=pgclient /usr/lib/postgresql/17/bin/pg_restore /usr/bin/pg_restore
+COPY --from=pgclient /usr/lib/postgresql/17/bin/psql /usr/bin/psql
+
+RUN apt-get update && \
+    apt-get install -y \
+        curl \
+        unzip \
+        && \
+    curl "https://awscli.amazonaws.com/awscli-exe-linux-x86_64.zip" -o "awscliv2.zip" && \
+    unzip awscliv2.zip && \
+    ./aws/install && \
+    rm -rf awscliv2.zip aws /var/lib/apt/lists/*
+
+ENV PYTHONDONTWRITEBYTECODE=1 \
+    PYTHONUNBUFFERED=1 \
+    AWS_DEFAULT_REGION=ca-central-1 \
+    AWS_RESPONSE_CHECKSUM_VALIDATION=when_required \
+    AWS_REQUEST_CHECKSUM_CALCULATION=when_required
+
+RUN useradd -m appuser && \
+    mkdir -p /app && \
+    chown -R appuser:appuser /app
+
+WORKDIR /app
+USER appuser
+
+# Copy restore script
+COPY --chown=appuser:appuser restore_database_from_s3.py .
+
+CMD ["python3", "restore_database_from_s3.py"]

backup_and_restore_database/restore_database/restore_database_from_s3.py

@@ -0,0 +1,74 @@
+import os
+import subprocess
+import sys
+from datetime import datetime
+import threading
+
+DB_HOST = os.environ["DB_HOST"]
+DB_PORT = os.environ["DB_PORT"]
+DB_NAME = os.environ["DB_NAME"]
+DB_USER = os.environ["DB_USER"]
+DB_PASSWORD = os.environ["DB_PASSWORD"]
+ENDPOINT_URL = os.environ["AWS_ENDPOINT_URL"]
+BACKUP_FILE = os.environ["BACKUP_FILE"]  # Format: "<date>-<database-name>.dump"
+BUCKET_NAME = os.environ["BUCKET_NAME"]
+
+S3_PATH = f"s3://{BUCKET_NAME}/{BACKUP_FILE}"
+os.environ["PGPASSWORD"] = DB_PASSWORD
+
+def log_stream(stream, prefix):
+        for line in stream:
+            print(f"{datetime.now()} [{prefix}] {line.decode().strip()}")
+
+aws_cp_cmd = [
+    "aws", "s3", "cp", S3_PATH, "-",
+    "--endpoint-url", ENDPOINT_URL,
+    "--expected-size", "60000000000"
+]
+
+pg_restore_cmd = [
+    "pg_restore",
+    "-h", DB_HOST,
+    "-p", DB_PORT,
+    "-U", DB_USER,
+    "-d", DB_NAME,
+    "-Fc",
+    "-v",
+    "--clean",
+    "--if-exists"
+]
+
+print(f"Starting restore from {S3_PATH} to {DB_NAME}...")
+
+try:
+    aws_proc = subprocess.Popen(aws_cp_cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+
+    restore_proc = subprocess.Popen(
+        pg_restore_cmd,
+        stdin=aws_proc.stdout,
+        stderr=subprocess.PIPE
+    )
+    aws_proc.stdout.close()
+
+
+    aws_thread = threading.Thread(target=log_stream, args=(aws_proc.stderr, "aws"))
+    restore_thread = threading.Thread(target=log_stream, args=(restore_proc.stderr, "pg_restore"))
+    aws_thread.start()
+    restore_thread.start()
+
+    aws_thread.join()
+    restore_thread.join()
+
+    if aws_proc.wait() != 0:
+        print("AWS download failed!")
+        sys.exit(1)
+
+    if restore_proc.wait() != 0:
+        print("pg_restore failed!")
+        sys.exit(1)
+
+    print("Restore completed successfully!")
+
+except Exception as e:
+    print(f"Restore failed: {str(e)}")
+    sys.exit(1)