Merge pull request #4690 from GSA-TTS/main

.github/workflows/deploy-application.yml

@@ -77,7 +77,7 @@ jobs:
           cf_password: ${{ secrets.CF_PASSWORD }}
           cf_org: gsa-tts-oros-fac
           cf_space: ${{ env.space }}
-          command: cf run-task gsa-fac -k 7G -m 3G --name deploy_backup --command "./fac-backup-util.sh v0.1.10 deploy_backup" --wait
+          command: cf run-task gsa-fac -k 7G -m 3G --name deploy_backup --command "./fac-backup-util.sh v0.1.11 deploy_backup" --wait
 
       - name: Deploy Preview to cloud.gov
         if: ${{ inputs.environment == 'preview' }}
@@ -124,5 +124,5 @@ jobs:
     secrets: inherit
     with:
       environment: ${{ inputs.environment }}
-      util_version: "v0.1.10"
+      util_version: "v0.1.11"
       backup_operation: "check_tables"

.github/workflows/fac-backup-scheduler.yml

@@ -19,6 +19,6 @@ jobs:
     secrets: inherit
     with:
       environment: ${{ matrix.environment.name }}
-      util_version: "v0.1.10"
+      util_version: "v0.1.11"
       backup_operation: "scheduled_backup"
 

.github/workflows/fac-backup-util-scheduled.yml

@@ -1,8 +1,8 @@
 ---
 name: Backup the database with fac-backup-utility
 ### Common Commands:
-# ./fac-backup-util.sh v0.1.10 scheduled_backup
-# ./fac-backup-util.sh v0.1.10 daily_backup
+# ./fac-backup-util.sh v0.1.11 scheduled_backup
+# ./fac-backup-util.sh v0.1.11 daily_backup
 on:
   workflow_call:
     inputs:

.github/workflows/fac-backup-util.yml

@@ -1,8 +1,8 @@
 ---
 name: Backup the database with fac-backup-utility
 ### Common Commands:
-# ./fac-backup-util.sh v0.1.10 initial_backup
-# ./fac-backup-util.sh v0.1.10 deploy_backup
+# ./fac-backup-util.sh v0.1.11 initial_backup
+# ./fac-backup-util.sh v0.1.11 deploy_backup
 on:
   workflow_dispatch:
     inputs:

.github/workflows/fac-check-tables-scheduler.yml

@@ -20,5 +20,5 @@ jobs:
     secrets: inherit
     with:
       environment: ${{ matrix.environment.name }}
-      util_version: "v0.1.10"
+      util_version: "v0.1.11"
       backup_operation: "check_tables"

.github/workflows/fac-check-tables.yml

@@ -1,7 +1,7 @@
 ---
 name: Check existing tables in an environment
 ### Common Commands:
-# ./fac-backup-util.sh v0.1.10 check_tables
+# ./fac-backup-util.sh v0.1.11 check_tables
 on:
   workflow_dispatch:
     inputs:

backend/config/settings.py

@@ -359,15 +359,16 @@
             )
 
         elif service["instance_name"] == "backups":
+            # Backups AWS S3 bucket for the app's backup files
             s3_creds = service["credentials"]
-            # Used for backing up the database https://django-dbbackup.readthedocs.io/en/master/storage.html#id2
-            DBBACKUP_STORAGE = "storages.backends.s3boto3.S3Boto3Storage"
-            DBBACKUP_STORAGE_OPTIONS = {
-                "access_key": s3_creds["access_key_id"],
-                "secret_key": s3_creds["secret_access_key"],
-                "bucket_name": s3_creds["bucket"],
-                "default_acl": "private",  # type: ignore
-            }
+
+            AWS_BACKUPS_ACCESS_KEY_ID = s3_creds["access_key_id"]
+            AWS_BACKUPS_SECRET_ACCESS_KEY = s3_creds["secret_access_key"]
+            AWS_BACKUPS_STORAGE_BUCKET_NAME = s3_creds["bucket"]
+            AWS_S3_BACKUPS_REGION_NAME = s3_creds["region"]
+            AWS_S3_BACKUPS_ENDPOINT = s3_creds["endpoint"]
+            AWS_S3_BACKUPS_ENDPOINT_URL = f"https://{AWS_S3_BACKUPS_ENDPOINT}"
+            AWS_PRIVATE_DEFAULT_ACL = "private"
 
     # secure headers
     MIDDLEWARE.append("csp.middleware.CSPMiddleware")

backend/curation/management/commands/fix_migrationinspectionrecord.py

@@ -0,0 +1,72 @@
+from django.core.management.base import BaseCommand
+from dissemination.models import MigrationInspectionRecord
+from config.settings import ENVIRONMENT, GSA_MIGRATION
+from django.db.models import Q
+
+import logging
+
+
+logger = logging.getLogger(__name__)
+logger.setLevel(logging.WARNING)
+
+
+class Command(BaseCommand):
+    help = """
+        Replace 'GSA_MIGRATION' with '' in policies_content and rate_content
+        of census_data in a note in dissemination_migrationinspectionrecord
+
+        Usage:
+        manage.py update_migrationinspectionrecord
+            --year <audit year>
+    """
+
+    def add_arguments(self, parser):
+        parser.add_argument(
+            "--year", help="Year(2016 through 2022)", type=str, default="2022"
+        )
+
+    def is_year_invalid(self, year):
+        valid_years = ["2016", "2017", "2018", "2019", "2020", "2021", "2022"]
+        return year not in valid_years
+
+    def handle(self, *args, **options):
+        if ENVIRONMENT not in [
+            "LOCAL",
+            "DEVELOPMENT",
+            "PREVIEW",
+            "STAGING",
+            "PRODUCTION",
+        ]:
+            print(f"Environment is not as expected, ENVIRONMENT={ENVIRONMENT}")
+            return
+
+        year = options.get("year")
+        if self.is_year_invalid(year):
+            print(
+                f"Invalid year {year}.  Expecting 2016 / 2017 / 2018 / 2019 / 2020 / 2021 / 2022"
+            )
+            return
+
+        migrationinspectionrecords = MigrationInspectionRecord.objects.filter(
+            Q(audit_year=year)
+        )
+        print(f"Count of {year} submissions: {len(migrationinspectionrecords)}")
+
+        count = 0
+        for migrationinspectionrecord in migrationinspectionrecords:
+            notes = []
+            is_updated = False
+            for note in migrationinspectionrecord.note:
+                if (
+                    note[0]["transformation_functions"][0]
+                    == "xform_missing_notes_records"
+                ) & (note[0]["census_data"][0]["value"] == GSA_MIGRATION):
+                    note[0]["census_data"][0]["value"] = ""
+                    is_updated = True
+                notes += [note]
+            if is_updated:
+                migrationinspectionrecord.note = notes
+                migrationinspectionrecord.save()
+                count += 1
+
+        print("Number of records updated = ", count)

backend/support/management/commands/delete_stale_backups.py

@@ -0,0 +1,80 @@
+# Usage:
+# Do a delete: python manage.py delete_stale_backups --days X --delete true
+# List objects: python manage.py delete_stale_backups --days X
+
+import boto3
+from datetime import datetime, timezone, timedelta
+from django.conf import settings
+from django.core.management.base import BaseCommand
+import sys
+
+
+class Command(BaseCommand):
+    def add_arguments(self, parser):
+        parser.add_argument(
+            "--days",
+            type=int,
+            required=True,
+            help="Max age a key(file) in days can have before we want to delete it. Value must be (14) or greater.",
+        )
+        parser.add_argument(
+            "--delete",
+            required=False,
+            default=False,
+            help="True/False. Actually do a delete. If not specified, just list the keys found that match.",
+        )
+
+    def handle(self, *args, **options):
+        days = options["days"]
+        delete = options["delete"]
+
+        if days < 14:
+            print(
+                "Days cannot less than 14 to prevent up-to-date backups from being deleted. Exiting..."
+            )
+            sys.exit(1)
+
+        s3_client = boto3.client(
+            "s3",
+            aws_access_key_id=settings.AWS_BACKUPS_ACCESS_KEY_ID,
+            aws_secret_access_key=settings.AWS_BACKUPS_SECRET_ACCESS_KEY,
+            endpoint_url=settings.AWS_S3_BACKUPS_ENDPOINT_URL,
+        )
+
+        paginator = s3_client.get_paginator("list_objects_v2")
+        pages = paginator.paginate(
+            Bucket=settings.AWS_BACKUPS_STORAGE_BUCKET_NAME, Prefix="backups/"
+        )
+
+        delete_older_than = datetime.now(timezone.utc) - timedelta(days=days)
+        total_count = 0
+        for page in pages:
+            if "Contents" in page:
+                for obj in page["Contents"]:
+
+                    # Get the last modified date of the object
+                    last_modified = obj["LastModified"]
+
+                    # If the object is older than one week, delete it
+                    # s3_client.delete_object(Bucket=settings.AWS_STORAGE_BUCKET_NAME, Key=f"backups/{item.file.name}")
+                    if delete:
+                        if last_modified < delete_older_than:
+                            print(
+                                f"Deleting {obj['Key']} last modified on {last_modified}"
+                            )
+                            s3_client.delete_object(
+                                Bucket=settings.AWS_BACKUPS_STORAGE_BUCKET_NAME,
+                                Key=obj["Key"],
+                            )
+                            total_count += 1
+                        else:
+                            print(
+                                f"Object {obj['Key']} younger than {delete_older_than}. Not deleting."
+                            )
+                    else:
+                        print(
+                            f"Delete not sent. {obj['Key']} was last modified on {last_modified}"
+                        )
+            else:
+                print("No objects found in the bucket.")
+        print(f"Total number of objects deleted: {total_count}")

docs/backups_and_restores.md

@@ -34,7 +34,7 @@ Information regarding the fac-backup-utility can be found [at the repository](ht
 Database backups occur in the following ways:
 1. An initial backup, where a backup has not been run in the target environment. This input of `initial_backup` is important, as when it does a the `db_to_db` command, it will not truncate the target table, as the table does not exist in the destination database.
 ```bash
-./fac-backup-util.sh v0.1.10 initial_backup
+./fac-backup-util.sh v0.1.11 initial_backup
 # Curl the utility
 # Install AWS
 # DB to S3 table dump (backups)
@@ -44,7 +44,7 @@ Database backups occur in the following ways:
 
 2. A deploy backup, where the `db_to_db` function is not called. This is a standard backup strategy before the application deploys, to ensure the s3 contents of the primary s3 are sync'd to the backups bucket, and a table dump is stored in the backups bucket.
 ```bash
-./fac-backup-util.sh v0.1.10 deploy_backup
+./fac-backup-util.sh v0.1.11 deploy_backup
 # Curl the utility
 # Install AWS
 # DB to S3 table dump (backups)
@@ -53,7 +53,7 @@ Database backups occur in the following ways:
 
 3. A scheduled backup is run every two hours, across each environment, ensuring that we have a clean backup in s3, rds, and the bucket contents are in sync.
 ```bash
-./fac-backup-util.sh v0.1.10 scheduled_backup
+./fac-backup-util.sh v0.1.11 scheduled_backup
 # Curl the utility
 # Install AWS
 # DB to S3 table dump (fac-db -> backups)
@@ -66,7 +66,7 @@ Restoring from backups can be run via workflow, from designated individuals. The
 
 1. S3 Restore takes a `operation-mm-DD-HH` input (ex `scheduled-06-04-10`), and is required for the backups to be restored. The utility looks in `s3://${bucket}/backups/operation-mm-DD-HH/` for its table dumps, and without supplying the target backups, it will not restore. Once it does a `--data-only` restoration, it will then sync the files from the backups bucket to the application bucket. We do this to ensure the contents of the application bucket are up to date, relative to the data in the database. We know that if we use the latest folder in `/backups/` then the contents of the s3 are the latest available, from the prior backup.
 ```bash
-./fac-restore-util.sh v0.1.10 s3_restore scheduled-06-04-10
+./fac-restore-util.sh v0.1.11 s3_restore scheduled-06-04-10
 # Curl the utility
 # Install AWS
 # DB to S3 table dump (backups -> fac-db) [Truncate target table before --data-only pg_restore]
@@ -81,7 +81,7 @@ daily-mm-dd
 
 2. Database to database restoration also can occur as well, using `psql` to dump the tables from the cold store database to the live database.
 ```bash
-./fac-restore-util.sh v0.1.10 db_restore
+./fac-restore-util.sh v0.1.11 db_restore
 # Curl the utility
 # Install AWS
 # DB to DB table dump (fac-snapshot-db -> fac-db) [Truncate target table before dump]

docs/deleting-backups.md

@@ -0,0 +1,53 @@
+### Informational document regarding Management Command "delete_stale_backups"
+
+The purpose of this document is to highlight examples for when a developer wishes to delete stale backups from the s3 bucket `backups`.
+
+**Warning:** This command is classified as a destructive command, and should only be run after receiving confirmation from members of the team, and after putting a formal annoucement in the developer slack channel. It is advised that after this command is run, to take a formal backup of the environment just for extra precautions.
+
+#### Information:
+The management command is located here: [delete_stale_backups.py](../backend/support/management/commands/delete_stale_backups.py). This command accepts two inputs. `--days` & `--delete`.
+- The value of `--days` must be greater than or equal to `14` (`--days 14`)
+- The value of `--delete` is required to actually perform the delete, and is a boolean (`--delete true`)
+- The full command to perform a delete will look like this:
+`python manage.py delete_stale_backups --days 14 --delete true`
+
+#### How to perform a delete
+1. Login to cloud.gov `cf login -a api.fr.cloud.gov --sso`
+2. Select the target environment if you have not done so after successful authentication `cf t -s <env>`
+3. Open a new terminal and tail the logs `cf logs gsa-fac | grep "delete_stale_backups"`
+4. Run the command via tasks:
+`cf run-task gsa-fac -k 2G -m 3G --name delete_stale_backups --command "python manage.py delete_stale_backups --days 14 --delete true" --wait`
+5. Wait for the command to finish.
+6. Navigate to [The backup environment action](https://github.com/GSA-TTS/FAC/actions/workflows/fac-backup-util.yml) and perform a backup with the following inputs or alternatively, navigate to [the scheduled backup action](https://github.com/GSA-TTS/FAC/actions/workflows/fac-backup-scheduler.yml) and run.
+```sh
+branch: main
+environment: <env where backups were just deleted (dev/staging/prod)>
+version: v0.1.11
+operation: on_demand_backup
+```
+
+#### Operation outputs examples (Fail):
+```
+~$ python manage.py delete_stale_backups --days 13
+Days cannot less than 14 to prevent up-to-date backups from being deleted. Exiting...
+~$
+
+~$ python manage.py delete_stale_backups --days 0 --delete true
+Days cannot less than 14 to prevent up-to-date backups from being deleted. Exiting...
+~$
+
+~$ python manage.py delete_stale_backups --days 14 --delete true
+Object backups/on-demand/02-04-13/public-audit_access.dump younger than 2025-01-22 18:44:02.406263+00:00. Not deleting.
+Object backups/on-demand/02-04-13/public-audit_deletedaccess.dump younger than 2025-01-22 18:44:02.406263+00:00. Not deleting.
+[...]
+```
+
+#### Operation outputs example (Pass):
+```
+~$ python manage.py delete_stale_backups --days 14 --delete true
+
+Deleting backups/on-demand/02-03-19/public-audit_access.dump last modified on 2025-01-22 18:44:02.406263+00:00
+Deleting backups/on-demand/02-03-19/public-audit_deletedaccess.dump last modified on 2025-01-22 18:44:02.406263+00:00
+[...]
+```
+

terraform/sandbox/sandbox.tf

@@ -8,6 +8,7 @@ module "sandbox" {
   login_client_id         = var.login_client_id
   login_secret_key        = var.login_secret_key
   branch_name             = var.branch_name
+  backups_s3_id           = module.sandbox-backups-bucket.bucket_id
 
   database_plan         = "medium-gp-psql"
   https_proxy_instances = 1
@@ -17,3 +18,13 @@ module "sandbox" {
     }
   )
 }
+
+module "sandbox-backups-bucket" {
+  source = "github.com/gsa-tts/terraform-cloudgov//s3?ref=v1.1.0"
+
+  cf_org_name   = var.cf_org_name
+  cf_space_name = "sandbox"
+  name          = "backups"
+  s3_plan_name  = "basic"
+  tags          = ["s3"]
+}

terraform/shared/modules/app/app.tf

@@ -74,6 +74,10 @@ resource "cloudfoundry_app" "fac_app" {
     service_instance = var.public_s3_id
   }
 
+  service_binding {
+    service_instance = var.backups_s3_id
+  }
+
   service_binding {
     service_instance = var.db_id
   }

terraform/shared/modules/app/variables.tf

@@ -48,6 +48,11 @@ variable "public_s3_id" {
   description = "the full string of the public s3 resource id"
 }
 
+variable "backups_s3_id" {
+  type        = string
+  description = "the full string of the backups s3 resource id"
+}
+
 variable "db_id" {
   type        = string
   description = "the full string of the core db resource id"

terraform/shared/modules/sandbox/app.tf

@@ -12,6 +12,7 @@ module "fac-app" {
   new_relic_creds_id      = cloudfoundry_user_provided_service.credentials.id
   private_s3_id           = module.s3-private.bucket_id
   public_s3_id            = module.s3-public.bucket_id
+  backups_s3_id           = var.backups_s3_id
   db_id                   = module.database.instance_id
   backup_db_id            = module.snapshot-database.instance_id
   app_instances           = 1