Merge pull request 'NX-10x: Reliability, error handling, runtime UX hardening, and migration safety gate (NX-101, NX-102, NX-103, NX-104)' (#32) from development into main

Reviewed-on: #32

.env.example

@@ -58,7 +58,3 @@ INIT_ADMIN_PASSWORD=ChangeMe123!
 # ------------------------------
 # Host port mapped to frontend container port 80.
 FRONTEND_PORT=5173
-# Base API URL used at frontend build time.
-# For reverse proxy + SSL, keep this relative to avoid mixed-content issues.
-# Example direct mode: VITE_API_URL=http://localhost:8000/api/v1
-VITE_API_URL=/api/v1

.github/workflows/docker-release.yml

@@ -0,0 +1,91 @@
+name: Docker Publish (Release)
+
+on:
+  release:
+    types: [published]
+  workflow_dispatch:
+    inputs:
+      version:
+        description: "Version tag to publish (e.g. 0.1.2 or v0.1.2)"
+        required: false
+        type: string
+
+jobs:
+  publish:
+    name: Build and Push Docker Images
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+
+    env:
+      # Optional repo variable. If unset, DOCKERHUB_USERNAME is used.
+      IMAGE_NAMESPACE: ${{ vars.DOCKERHUB_NAMESPACE }}
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Resolve version/tag
+        id: ver
+        shell: bash
+        run: |
+          RAW_TAG="${{ github.event.release.tag_name }}"
+          if [ -z "$RAW_TAG" ]; then
+            RAW_TAG="${{ inputs.version }}"
+          fi
+          if [ -z "$RAW_TAG" ]; then
+            RAW_TAG="${GITHUB_REF_NAME}"
+          fi
+
+          CLEAN_TAG="${RAW_TAG#v}"
+          echo "raw=$RAW_TAG" >> "$GITHUB_OUTPUT"
+          echo "clean=$CLEAN_TAG" >> "$GITHUB_OUTPUT"
+
+      - name: Set image namespace
+        id: ns
+        shell: bash
+        run: |
+          NS="${IMAGE_NAMESPACE}"
+          if [ -z "$NS" ]; then
+            NS="${{ secrets.DOCKERHUB_USERNAME }}"
+          fi
+          if [ -z "$NS" ]; then
+            echo "Missing Docker Hub namespace. Set repo var DOCKERHUB_NAMESPACE or secret DOCKERHUB_USERNAME."
+            exit 1
+          fi
+          echo "value=$NS" >> "$GITHUB_OUTPUT"
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+
+      - name: Login to Docker Hub
+        uses: docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKERHUB_USERNAME }}
+          password: ${{ secrets.DOCKERHUB_TOKEN }}
+
+      - name: Build and push backend image
+        uses: docker/build-push-action@v6
+        with:
+          context: ./backend
+          file: ./backend/Dockerfile
+          push: true
+          tags: |
+            ${{ steps.ns.outputs.value }}/nexapg-backend:${{ steps.ver.outputs.clean }}
+            ${{ steps.ns.outputs.value }}/nexapg-backend:latest
+          cache-from: type=registry,ref=${{ steps.ns.outputs.value }}/nexapg-backend:buildcache
+          cache-to: type=registry,ref=${{ steps.ns.outputs.value }}/nexapg-backend:buildcache,mode=max
+
+      - name: Build and push frontend image
+        uses: docker/build-push-action@v6
+        with:
+          context: ./frontend
+          file: ./frontend/Dockerfile
+          push: true
+          build-args: |
+            VITE_API_URL=/api/v1
+          tags: |
+            ${{ steps.ns.outputs.value }}/nexapg-frontend:${{ steps.ver.outputs.clean }}
+            ${{ steps.ns.outputs.value }}/nexapg-frontend:latest
+          cache-from: type=registry,ref=${{ steps.ns.outputs.value }}/nexapg-frontend:buildcache
+          cache-to: type=registry,ref=${{ steps.ns.outputs.value }}/nexapg-frontend:buildcache,mode=max

.github/workflows/migration-safety.yml

@@ -0,0 +1,86 @@
+name: Migration Safety
+
+on:
+  push:
+    branches: ["main", "master"]
+  pull_request:
+
+jobs:
+  migration-safety:
+    name: Alembic upgrade/downgrade safety
+    runs-on: ubuntu-latest
+
+    services:
+      postgres:
+        image: postgres:16
+        env:
+          POSTGRES_DB: nexapg
+          POSTGRES_USER: nexapg
+          POSTGRES_PASSWORD: nexapg
+        ports:
+          - 5432:5432
+        options: >-
+          --health-cmd "pg_isready -U nexapg -d nexapg"
+          --health-interval 5s
+          --health-timeout 5s
+          --health-retries 30
+
+    env:
+      DB_HOST: postgres
+      DB_PORT: 5432
+      DB_NAME: nexapg
+      DB_USER: nexapg
+      DB_PASSWORD: nexapg
+      JWT_SECRET_KEY: ci-jwt-secret-key
+      ENCRYPTION_KEY: MDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDA=
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+
+      - name: Install backend dependencies
+        run: pip install -r backend/requirements.txt
+
+      - name: Install PostgreSQL client tools
+        run: sudo apt-get update && sudo apt-get install -y postgresql-client
+
+      - name: Wait for PostgreSQL
+        env:
+          PGPASSWORD: nexapg
+        run: |
+          for i in $(seq 1 60); do
+            if pg_isready -h postgres -p 5432 -U nexapg -d nexapg; then
+              exit 0
+            fi
+            sleep 2
+          done
+          echo "PostgreSQL did not become ready in time."
+          exit 1
+
+      - name: Alembic upgrade -> downgrade -1 -> upgrade
+        working-directory: backend
+        run: |
+          alembic upgrade head
+          alembic downgrade -1
+          alembic upgrade head
+
+      - name: Validate schema consistency after roundtrip
+        env:
+          PGPASSWORD: nexapg
+        run: |
+          cd backend
+          alembic upgrade head
+          pg_dump -h postgres -p 5432 -U nexapg -d nexapg --schema-only --no-owner --no-privileges \
+            | sed '/^\\restrict /d; /^\\unrestrict /d' > /tmp/schema_head_before.sql
+
+          alembic downgrade -1
+          alembic upgrade head
+          pg_dump -h postgres -p 5432 -U nexapg -d nexapg --schema-only --no-owner --no-privileges \
+            | sed '/^\\restrict /d; /^\\unrestrict /d' > /tmp/schema_head_after.sql
+
+          diff -u /tmp/schema_head_before.sql /tmp/schema_head_after.sql

Makefile

@@ -1,7 +1,8 @@
 .PHONY: up down logs migrate
 
 up:
-	docker compose up -d --build
+	docker compose pull
+	docker compose up -d
 
 down:
 	docker compose down

README.md

@@ -9,7 +9,7 @@ It combines FastAPI, React, and PostgreSQL in a Docker Compose stack with RBAC,
 
 ## Table of Contents
 
-- [Quick Start](#quick-start)
+- [Quick Deploy (Prebuilt Images)](#quick-deploy-prebuilt-images)
 - [Prerequisites](#prerequisites)
 - [Make Commands](#make-commands)
 - [Configuration Reference (`.env`)](#configuration-reference-env)
@@ -17,6 +17,7 @@ It combines FastAPI, React, and PostgreSQL in a Docker Compose stack with RBAC,
 - [Service Information](#service-information)
 - [Target Owner Notifications](#target-owner-notifications)
 - [API Overview](#api-overview)
+- [API Error Format](#api-error-format)
 - [`pg_stat_statements` Requirement](#pg_stat_statements-requirement)
 - [Reverse Proxy / SSL Guidance](#reverse-proxy--ssl-guidance)
 - [PostgreSQL Compatibility Smoke Test](#postgresql-compatibility-smoke-test)
@@ -93,27 +94,50 @@ Optional:
 
 - `psql` for manual DB checks
 
-## Quick Start
+## Quick Deploy (Prebuilt Images)
 
-1. Copy environment template:
+If you only want to run NexaPG from published Docker Hub images, use the bootstrap script:
 
 ```bash
-cp .env.example .env
+mkdir -p /opt/NexaPG
+cd /opt/NexaPG
+wget -O bootstrap-compose.sh https://git.nesterovic.cc/nessi/NexaPG/raw/branch/main/ops/scripts/bootstrap-compose.sh
+chmod +x bootstrap-compose.sh
+./bootstrap-compose.sh
 ```
 
-2. Generate a Fernet key and set `ENCRYPTION_KEY` in `.env`:
+This downloads:
+
+- `docker-compose.yml`
+- `.env.example`
+- `Makefile`
+
+Then:
 
 ```bash
+# generate JWT secret
+python -c "import secrets; print(secrets.token_urlsafe(64))"
+# generate Fernet encryption key
 python -c "from cryptography.fernet import Fernet; print(Fernet.generate_key().decode())"
-```
+# put both values into .env (JWT_SECRET_KEY / ENCRYPTION_KEY)
-
+# note: .env is auto-created by bootstrap if it does not exist
-3. Start the stack:
-
-```bash
 make up
 ```
 
-4. Open the application:
+Manual download alternative:
+
+```bash
+mkdir -p /opt/NexaPG
+cd /opt/NexaPG
+wget https://git.nesterovic.cc/nessi/NexaPG/raw/branch/main/docker-compose.yml
+wget https://git.nesterovic.cc/nessi/NexaPG/raw/branch/main/.env.example
+wget https://git.nesterovic.cc/nessi/NexaPG/raw/branch/main/Makefile
+cp .env.example .env
+```
+
+`make up` pulls `nesterovicit/nexapg-backend:latest` and `nesterovicit/nexapg-frontend:latest`, then starts the stack.
+
+Open the application:
 
 - Frontend: `http://<SERVER_IP>:<FRONTEND_PORT>`
 - API base: `http://<SERVER_IP>:<BACKEND_PORT>/api/v1`
@@ -127,7 +151,7 @@ Initial admin bootstrap user (created from `.env` if missing):
 ## Make Commands
 
 ```bash
-make up        # build and start all services
+make up        # pull latest images and start all services
 make down      # stop all services
 make logs      # follow compose logs
 make migrate   # optional/manual: run alembic upgrade head in backend container
@@ -183,12 +207,6 @@ Note: Migrations run automatically when the backend container starts (`entrypoin
 | Variable | Description |
 |---|---|
 | `FRONTEND_PORT` | Host port mapped to frontend container port `80` |
-| `VITE_API_URL` | Frontend API base URL (build-time) |
-
-Recommended values for `VITE_API_URL`:
-
-- Reverse proxy setup: `/api/v1`
-- Direct backend access: `http://<SERVER_IP>:<BACKEND_PORT>/api/v1`
 
 ## Core Functional Areas
 
@@ -302,6 +320,37 @@ Email alert routing is target-specific:
 - `GET /api/v1/service/info`
 - `POST /api/v1/service/info/check`
 
+## API Error Format
+
+All 4xx/5xx responses use a consistent JSON payload:
+
+```json
+{
+  "code": "validation_error",
+  "message": "Request validation failed",
+  "details": [],
+  "request_id": "c8f0f888-2365-4b86-a5de-b3f0e9df4a4b"
+}
+```
+
+Common fields:
+
+- `code`: stable machine-readable error code
+- `message`: human-readable summary
+- `details`: optional extra context (validation list, debug context, etc.)
+- `request_id`: request correlation ID (also returned in `X-Request-ID` header)
+
+Common error codes:
+
+- `bad_request` (`400`)
+- `unauthorized` (`401`)
+- `forbidden` (`403`)
+- `not_found` (`404`)
+- `conflict` (`409`)
+- `validation_error` (`422`)
+- `target_unreachable` (`503`)
+- `internal_error` (`500`)
+
 ## `pg_stat_statements` Requirement
 
 Query Insights requires `pg_stat_statements` on the monitored target:
@@ -318,7 +367,7 @@ For production, serve frontend and API under the same public origin via reverse
 
 - Frontend URL example: `https://monitor.example.com`
 - Proxy API path `/api/` to backend service
-- Use `VITE_API_URL=/api/v1`
+- Route `/api/v1` to the backend service
 
 This prevents mixed-content and CORS issues.
 
@@ -351,8 +400,7 @@ docker compose logs --tail=200 db
 
 ### CORS or mixed-content issues behind SSL proxy
 
-- Set `VITE_API_URL=/api/v1`
+- Ensure proxy forwards `/api/` (or `/api/v1`) to backend
-- Ensure proxy forwards `/api/` to backend
 - Set correct frontend origin(s) in `CORS_ORIGINS`
 
 ### `rejected SSL upgrade` for a target

backend/Dockerfile

@@ -1,4 +1,4 @@
-FROM python:3.12-slim AS base
+FROM python:3.13-slim AS base
 
 ENV PYTHONDONTWRITEBYTECODE=1
 ENV PYTHONUNBUFFERED=1
@@ -6,6 +6,10 @@ ENV PIP_NO_CACHE_DIR=1
 
 WORKDIR /app
 
+RUN apt-get update \
+    && apt-get upgrade -y \
+    && rm -rf /var/lib/apt/lists/*
+
 RUN addgroup --system app && adduser --system --ingroup app app
 
 COPY requirements.txt /app/requirements.txt

backend/alembic/versions/0009_user_profile_fields.py

@@ -0,0 +1,26 @@
+"""add user first and last name fields
+
+Revision ID: 0009_user_profile_fields
+Revises: 0008_service_settings
+Create Date: 2026-02-13
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+revision = "0009_user_profile_fields"
+down_revision = "0008_service_settings"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.add_column("users", sa.Column("first_name", sa.String(length=120), nullable=True))
+    op.add_column("users", sa.Column("last_name", sa.String(length=120), nullable=True))
+
+
+def downgrade() -> None:
+    op.drop_column("users", "last_name")
+    op.drop_column("users", "first_name")
+

backend/app/api/routes/admin_settings.py

@@ -9,6 +9,7 @@ from sqlalchemy.ext.asyncio import AsyncSession
 
 from app.core.db import get_db
 from app.core.deps import require_roles
+from app.core.errors import api_error
 from app.models.models import EmailNotificationSettings, User
 from app.schemas.admin_settings import EmailSettingsOut, EmailSettingsTestRequest, EmailSettingsUpdate
 from app.services.audit import write_audit_log
@@ -96,9 +97,9 @@ async def test_email_settings(
 ) -> dict:
     settings = await _get_or_create_settings(db)
     if not settings.smtp_host:
-        raise HTTPException(status_code=400, detail="SMTP host is not configured")
+        raise HTTPException(status_code=400, detail=api_error("smtp_host_missing", "SMTP host is not configured"))
     if not settings.from_email:
-        raise HTTPException(status_code=400, detail="From email is not configured")
+        raise HTTPException(status_code=400, detail=api_error("smtp_from_email_missing", "From email is not configured"))
 
     password = decrypt_secret(settings.encrypted_smtp_password) if settings.encrypted_smtp_password else None
     message = EmailMessage()
@@ -126,7 +127,10 @@ async def test_email_settings(
                     smtp.login(settings.smtp_username, password or "")
                 smtp.send_message(message)
     except Exception as exc:
-        raise HTTPException(status_code=400, detail=f"SMTP test failed: {exc}")
+        raise HTTPException(
+            status_code=400,
+            detail=api_error("smtp_test_failed", "SMTP test failed", {"error": str(exc)}),
+        ) from exc
 
     await write_audit_log(db, "admin.email_settings.test", admin.id, {"recipient": str(payload.recipient)})
     return {"status": "sent", "recipient": str(payload.recipient)}

backend/app/api/routes/admin_users.py

@@ -3,6 +3,7 @@ from sqlalchemy import select
 from sqlalchemy.ext.asyncio import AsyncSession
 from app.core.db import get_db
 from app.core.deps import require_roles
+from app.core.errors import api_error
 from app.core.security import hash_password
 from app.models.models import User
 from app.schemas.user import UserCreate, UserOut, UserUpdate
@@ -22,8 +23,14 @@ async def list_users(admin: User = Depends(require_roles("admin")), db: AsyncSes
 async def create_user(payload: UserCreate, admin: User = Depends(require_roles("admin")), db: AsyncSession = Depends(get_db)) -> UserOut:
     exists = await db.scalar(select(User).where(User.email == payload.email))
     if exists:
-        raise HTTPException(status_code=409, detail="Email already exists")
+        raise HTTPException(status_code=409, detail=api_error("email_exists", "Email already exists"))
-    user = User(email=payload.email, password_hash=hash_password(payload.password), role=payload.role)
+    user = User(
+        email=payload.email,
+        first_name=payload.first_name,
+        last_name=payload.last_name,
+        password_hash=hash_password(payload.password),
+        role=payload.role,
+    )
     db.add(user)
     await db.commit()
     await db.refresh(user)
@@ -40,10 +47,17 @@ async def update_user(
 ) -> UserOut:
     user = await db.scalar(select(User).where(User.id == user_id))
     if not user:
-        raise HTTPException(status_code=404, detail="User not found")
+        raise HTTPException(status_code=404, detail=api_error("user_not_found", "User not found"))
     update_data = payload.model_dump(exclude_unset=True)
-    if "password" in update_data and update_data["password"]:
+    next_email = update_data.get("email")
-        user.password_hash = hash_password(update_data.pop("password"))
+    if next_email and next_email != user.email:
+        existing = await db.scalar(select(User).where(User.email == next_email))
+        if existing and existing.id != user.id:
+            raise HTTPException(status_code=409, detail=api_error("email_exists", "Email already exists"))
+    if "password" in update_data:
+        raw_password = update_data.pop("password")
+        if raw_password:
+            user.password_hash = hash_password(raw_password)
     for key, value in update_data.items():
         setattr(user, key, value)
     await db.commit()
@@ -55,10 +69,10 @@ async def update_user(
 @router.delete("/{user_id}")
 async def delete_user(user_id: int, admin: User = Depends(require_roles("admin")), db: AsyncSession = Depends(get_db)) -> dict:
     if user_id == admin.id:
-        raise HTTPException(status_code=400, detail="Cannot delete yourself")
+        raise HTTPException(status_code=400, detail=api_error("cannot_delete_self", "Cannot delete yourself"))
     user = await db.scalar(select(User).where(User.id == user_id))
     if not user:
-        raise HTTPException(status_code=404, detail="User not found")
+        raise HTTPException(status_code=404, detail=api_error("user_not_found", "User not found"))
     await db.delete(user)
     await db.commit()
     await write_audit_log(db, "admin.user.delete", admin.id, {"deleted_user_id": user_id})

backend/app/api/routes/alerts.py

@@ -4,6 +4,7 @@ from sqlalchemy.ext.asyncio import AsyncSession
 
 from app.core.db import get_db
 from app.core.deps import get_current_user, require_roles
+from app.core.errors import api_error
 from app.models.models import AlertDefinition, Target, User
 from app.schemas.alert import (
     AlertDefinitionCreate,
@@ -33,7 +34,7 @@ async def _validate_target_exists(db: AsyncSession, target_id: int | None) -> No
         return
     target_exists = await db.scalar(select(Target.id).where(Target.id == target_id))
     if target_exists is None:
-        raise HTTPException(status_code=404, detail="Target not found")
+        raise HTTPException(status_code=404, detail=api_error("target_not_found", "Target not found"))
 
 
 @router.get("/status", response_model=AlertStatusResponse)
@@ -101,7 +102,7 @@ async def update_alert_definition(
 ) -> AlertDefinitionOut:
     definition = await db.scalar(select(AlertDefinition).where(AlertDefinition.id == definition_id))
     if definition is None:
-        raise HTTPException(status_code=404, detail="Alert definition not found")
+        raise HTTPException(status_code=404, detail=api_error("alert_definition_not_found", "Alert definition not found"))
 
     updates = payload.model_dump(exclude_unset=True)
     if "target_id" in updates:
@@ -131,7 +132,7 @@ async def delete_alert_definition(
 ) -> dict:
     definition = await db.scalar(select(AlertDefinition).where(AlertDefinition.id == definition_id))
     if definition is None:
-        raise HTTPException(status_code=404, detail="Alert definition not found")
+        raise HTTPException(status_code=404, detail=api_error("alert_definition_not_found", "Alert definition not found"))
     await db.delete(definition)
     await db.commit()
     invalidate_alert_cache()
@@ -148,7 +149,7 @@ async def test_alert_definition(
     _ = user
     target = await db.scalar(select(Target).where(Target.id == payload.target_id))
     if target is None:
-        raise HTTPException(status_code=404, detail="Target not found")
+        raise HTTPException(status_code=404, detail=api_error("target_not_found", "Target not found"))
     try:
         value = await run_scalar_sql_for_target(target, payload.sql_text)
         return AlertDefinitionTestResponse(ok=True, value=value)

backend/app/api/routes/auth.py

@@ -1,10 +1,11 @@
 from fastapi import APIRouter, Depends, HTTPException, status
-from jose import JWTError, jwt
+import jwt
 from sqlalchemy import select
 from sqlalchemy.ext.asyncio import AsyncSession
 from app.core.config import get_settings
 from app.core.db import get_db
 from app.core.deps import get_current_user
+from app.core.errors import api_error
 from app.core.security import create_access_token, create_refresh_token, verify_password
 from app.models.models import User
 from app.schemas.auth import LoginRequest, RefreshRequest, TokenResponse
@@ -19,7 +20,10 @@ settings = get_settings()
 async def login(payload: LoginRequest, db: AsyncSession = Depends(get_db)) -> TokenResponse:
     user = await db.scalar(select(User).where(User.email == payload.email))
     if not user or not verify_password(payload.password, user.password_hash):
-        raise HTTPException(status_code=status.HTTP_401_UNAUTHORIZED, detail="Invalid credentials")
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=api_error("invalid_credentials", "Invalid credentials"),
+        )
 
     await write_audit_log(db, action="auth.login", user_id=user.id, payload={"email": user.email})
     return TokenResponse(access_token=create_access_token(str(user.id)), refresh_token=create_refresh_token(str(user.id)))
@@ -29,15 +33,24 @@ async def login(payload: LoginRequest, db: AsyncSession = Depends(get_db)) -> To
 async def refresh(payload: RefreshRequest, db: AsyncSession = Depends(get_db)) -> TokenResponse:
     try:
         token_payload = jwt.decode(payload.refresh_token, settings.jwt_secret_key, algorithms=[settings.jwt_algorithm])
-    except JWTError as exc:
+    except jwt.InvalidTokenError as exc:
-        raise HTTPException(status_code=status.HTTP_401_UNAUTHORIZED, detail="Invalid refresh token") from exc
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=api_error("invalid_refresh_token", "Invalid refresh token"),
+        ) from exc
 
     if token_payload.get("type") != "refresh":
-        raise HTTPException(status_code=status.HTTP_401_UNAUTHORIZED, detail="Invalid refresh token type")
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=api_error("invalid_refresh_token_type", "Invalid refresh token type"),
+        )
     user_id = token_payload.get("sub")
     user = await db.scalar(select(User).where(User.id == int(user_id)))
     if not user:
-        raise HTTPException(status_code=status.HTTP_401_UNAUTHORIZED, detail="User not found")
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=api_error("user_not_found", "User not found"),
+        )
 
     await write_audit_log(db, action="auth.refresh", user_id=user.id, payload={})
     return TokenResponse(access_token=create_access_token(str(user.id)), refresh_token=create_refresh_token(str(user.id)))

backend/app/api/routes/me.py

@@ -2,6 +2,7 @@ from fastapi import APIRouter, Depends, HTTPException, status
 from sqlalchemy.ext.asyncio import AsyncSession
 from app.core.db import get_db
 from app.core.deps import get_current_user
+from app.core.errors import api_error
 from app.core.security import hash_password, verify_password
 from app.models.models import User
 from app.schemas.user import UserOut, UserPasswordChange
@@ -22,10 +23,16 @@ async def change_password(
     db: AsyncSession = Depends(get_db),
 ) -> dict:
     if not verify_password(payload.current_password, user.password_hash):
-        raise HTTPException(status_code=status.HTTP_400_BAD_REQUEST, detail="Current password is incorrect")
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=api_error("invalid_current_password", "Current password is incorrect"),
+        )
 
     if verify_password(payload.new_password, user.password_hash):
-        raise HTTPException(status_code=status.HTTP_400_BAD_REQUEST, detail="New password must be different")
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=api_error("password_reuse_not_allowed", "New password must be different"),
+        )
 
     user.password_hash = hash_password(payload.new_password)
     await db.commit()

backend/app/api/routes/targets.py

@@ -8,6 +8,7 @@ from sqlalchemy.ext.asyncio import AsyncSession
 
 from app.core.db import get_db
 from app.core.deps import get_current_user, require_roles
+from app.core.errors import api_error
 from app.models.models import Metric, QueryStat, Target, TargetOwner, User
 from app.schemas.metric import MetricOut, QueryStatOut
 from app.schemas.overview import DatabaseOverviewOut
@@ -85,7 +86,10 @@ async def _discover_databases(payload: TargetCreate) -> list[str]:
         )
         return [row["datname"] for row in rows if row["datname"]]
     except Exception as exc:
-        raise HTTPException(status_code=400, detail=f"Database discovery failed: {exc}")
+        raise HTTPException(
+            status_code=400,
+            detail=api_error("database_discovery_failed", "Database discovery failed", {"error": str(exc)}),
+        )
     finally:
         if conn:
             await conn.close()
@@ -131,7 +135,10 @@ async def test_target_connection(
         version = await conn.fetchval("SHOW server_version")
         return {"ok": True, "message": "Connection successful", "server_version": version}
     except Exception as exc:
-        raise HTTPException(status_code=400, detail=f"Connection failed: {exc}")
+        raise HTTPException(
+            status_code=400,
+            detail=api_error("connection_test_failed", "Connection test failed", {"error": str(exc)}),
+        )
     finally:
         if conn:
             await conn.close()
@@ -147,7 +154,10 @@ async def create_target(
     if owner_ids:
         owners_exist = (await db.scalars(select(User.id).where(User.id.in_(owner_ids)))).all()
         if len(set(owners_exist)) != len(owner_ids):
-            raise HTTPException(status_code=400, detail="One or more owner users were not found")
+            raise HTTPException(
+                status_code=400,
+                detail=api_error("owner_users_not_found", "One or more owner users were not found"),
+            )
 
     encrypted_password = encrypt_secret(payload.password)
     created_targets: list[Target] = []
@@ -155,7 +165,10 @@ async def create_target(
     if payload.discover_all_databases:
         databases = await _discover_databases(payload)
         if not databases:
-            raise HTTPException(status_code=400, detail="No databases discovered on target")
+            raise HTTPException(
+                status_code=400,
+                detail=api_error("no_databases_discovered", "No databases discovered on target"),
+            )
         group_id = str(uuid4())
         base_tags = payload.tags or {}
         for dbname in databases:
@@ -194,7 +207,10 @@ async def create_target(
                 await _set_target_owners(db, target.id, owner_ids, user.id)
 
         if not created_targets:
-            raise HTTPException(status_code=400, detail="All discovered databases already exist as targets")
+            raise HTTPException(
+                status_code=400,
+                detail=api_error("all_discovered_databases_exist", "All discovered databases already exist as targets"),
+            )
         await db.commit()
         for item in created_targets:
             await db.refresh(item)
@@ -247,7 +263,7 @@ async def get_target(target_id: int, user: User = Depends(get_current_user), db:
     _ = user
     target = await db.scalar(select(Target).where(Target.id == target_id))
     if not target:
-        raise HTTPException(status_code=404, detail="Target not found")
+        raise HTTPException(status_code=404, detail=api_error("target_not_found", "Target not found"))
     owner_map = await _owners_by_target_ids(db, [target.id])
     return _target_out_with_owners(target, owner_map.get(target.id, []))
 
@@ -261,7 +277,7 @@ async def update_target(
 ) -> TargetOut:
     target = await db.scalar(select(Target).where(Target.id == target_id))
     if not target:
-        raise HTTPException(status_code=404, detail="Target not found")
+        raise HTTPException(status_code=404, detail=api_error("target_not_found", "Target not found"))
 
     updates = payload.model_dump(exclude_unset=True)
     owner_user_ids = updates.pop("owner_user_ids", None)
@@ -273,7 +289,10 @@ async def update_target(
     if owner_user_ids is not None:
         owners_exist = (await db.scalars(select(User.id).where(User.id.in_(owner_user_ids)))).all()
         if len(set(owners_exist)) != len(set(owner_user_ids)):
-            raise HTTPException(status_code=400, detail="One or more owner users were not found")
+            raise HTTPException(
+                status_code=400,
+                detail=api_error("owner_users_not_found", "One or more owner users were not found"),
+            )
         await _set_target_owners(db, target.id, owner_user_ids, user.id)
 
     await db.commit()
@@ -292,12 +311,15 @@ async def set_target_owners(
 ) -> list[TargetOwnerOut]:
     target = await db.scalar(select(Target).where(Target.id == target_id))
     if not target:
-        raise HTTPException(status_code=404, detail="Target not found")
+        raise HTTPException(status_code=404, detail=api_error("target_not_found", "Target not found"))
     owner_user_ids = sorted(set(payload.user_ids))
     if owner_user_ids:
         owners_exist = (await db.scalars(select(User.id).where(User.id.in_(owner_user_ids)))).all()
         if len(set(owners_exist)) != len(set(owner_user_ids)):
-            raise HTTPException(status_code=400, detail="One or more owner users were not found")
+            raise HTTPException(
+                status_code=400,
+                detail=api_error("owner_users_not_found", "One or more owner users were not found"),
+            )
     await _set_target_owners(db, target_id, owner_user_ids, user.id)
     await db.commit()
     await write_audit_log(db, "target.owners.update", user.id, {"target_id": target_id, "owner_user_ids": owner_user_ids})
@@ -321,7 +343,7 @@ async def get_target_owners(
     _ = user
     target = await db.scalar(select(Target).where(Target.id == target_id))
     if not target:
-        raise HTTPException(status_code=404, detail="Target not found")
+        raise HTTPException(status_code=404, detail=api_error("target_not_found", "Target not found"))
     rows = (
         await db.execute(
             select(User.id, User.email, User.role)
@@ -341,7 +363,7 @@ async def delete_target(
 ) -> dict:
     target = await db.scalar(select(Target).where(Target.id == target_id))
     if not target:
-        raise HTTPException(status_code=404, detail="Target not found")
+        raise HTTPException(status_code=404, detail=api_error("target_not_found", "Target not found"))
     await db.delete(target)
     await db.commit()
     await write_audit_log(db, "target.delete", user.id, {"target_id": target_id})
@@ -369,7 +391,22 @@ async def get_metrics(
 
 
 async def _live_conn(target: Target) -> asyncpg.Connection:
-    return await asyncpg.connect(dsn=build_target_dsn(target))
+    try:
+        return await asyncpg.connect(dsn=build_target_dsn(target))
+    except (OSError, asyncpg.PostgresError) as exc:
+        raise HTTPException(
+            status_code=503,
+            detail=api_error(
+                "target_unreachable",
+                "Target database is not reachable",
+                {
+                    "target_id": target.id,
+                    "host": target.host,
+                    "port": target.port,
+                    "error": str(exc),
+                },
+            ),
+        ) from exc
 
 
 @router.get("/{target_id}/locks")
@@ -377,7 +414,7 @@ async def get_locks(target_id: int, user: User = Depends(get_current_user), db:
     _ = user
     target = await db.scalar(select(Target).where(Target.id == target_id))
     if not target:
-        raise HTTPException(status_code=404, detail="Target not found")
+        raise HTTPException(status_code=404, detail=api_error("target_not_found", "Target not found"))
     conn = await _live_conn(target)
     try:
         rows = await conn.fetch(
@@ -398,7 +435,7 @@ async def get_activity(target_id: int, user: User = Depends(get_current_user), d
     _ = user
     target = await db.scalar(select(Target).where(Target.id == target_id))
     if not target:
-        raise HTTPException(status_code=404, detail="Target not found")
+        raise HTTPException(status_code=404, detail=api_error("target_not_found", "Target not found"))
     conn = await _live_conn(target)
     try:
         rows = await conn.fetch(
@@ -420,7 +457,7 @@ async def get_top_queries(target_id: int, user: User = Depends(get_current_user)
     _ = user
     target = await db.scalar(select(Target).where(Target.id == target_id))
     if not target:
-        raise HTTPException(status_code=404, detail="Target not found")
+        raise HTTPException(status_code=404, detail=api_error("target_not_found", "Target not found"))
     if not target.use_pg_stat_statements:
         return []
     rows = (
@@ -450,5 +487,20 @@ async def get_overview(target_id: int, user: User = Depends(get_current_user), d
     _ = user
     target = await db.scalar(select(Target).where(Target.id == target_id))
     if not target:
-        raise HTTPException(status_code=404, detail="Target not found")
+        raise HTTPException(status_code=404, detail=api_error("target_not_found", "Target not found"))
-    return await get_target_overview(target)
+    try:
+        return await get_target_overview(target)
+    except (OSError, asyncpg.PostgresError) as exc:
+        raise HTTPException(
+            status_code=503,
+            detail=api_error(
+                "target_unreachable",
+                "Target database is not reachable",
+                {
+                    "target_id": target.id,
+                    "host": target.host,
+                    "port": target.port,
+                    "error": str(exc),
+                },
+            ),
+        ) from exc

backend/app/core/config.py

@@ -2,7 +2,7 @@ from functools import lru_cache
 from pydantic import field_validator
 from pydantic_settings import BaseSettings, SettingsConfigDict
 
-NEXAPG_VERSION = "0.1.2"
+NEXAPG_VERSION = "0.2.0"
 
 
 class Settings(BaseSettings):

backend/app/core/deps.py

@@ -1,10 +1,11 @@
 from fastapi import Depends, HTTPException, status
 from fastapi.security import HTTPAuthorizationCredentials, HTTPBearer
-from jose import JWTError, jwt
+import jwt
 from sqlalchemy import select
 from sqlalchemy.ext.asyncio import AsyncSession
 from app.core.config import get_settings
 from app.core.db import get_db
+from app.core.errors import api_error
 from app.models.models import User
 
 settings = get_settings()
@@ -16,27 +17,42 @@ async def get_current_user(
     db: AsyncSession = Depends(get_db),
 ) -> User:
     if not credentials:
-        raise HTTPException(status_code=status.HTTP_401_UNAUTHORIZED, detail="Missing token")
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=api_error("missing_token", "Missing token"),
+        )
     token = credentials.credentials
     try:
         payload = jwt.decode(token, settings.jwt_secret_key, algorithms=[settings.jwt_algorithm])
-    except JWTError as exc:
+    except jwt.InvalidTokenError as exc:
-        raise HTTPException(status_code=status.HTTP_401_UNAUTHORIZED, detail="Invalid token") from exc
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=api_error("invalid_token", "Invalid token"),
+        ) from exc
 
     if payload.get("type") != "access":
-        raise HTTPException(status_code=status.HTTP_401_UNAUTHORIZED, detail="Invalid token type")
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=api_error("invalid_token_type", "Invalid token type"),
+        )
 
     user_id = payload.get("sub")
     user = await db.scalar(select(User).where(User.id == int(user_id)))
     if not user:
-        raise HTTPException(status_code=status.HTTP_401_UNAUTHORIZED, detail="User not found")
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=api_error("user_not_found", "User not found"),
+        )
     return user
 
 
 def require_roles(*roles: str):
     async def role_dependency(user: User = Depends(get_current_user)) -> User:
         if user.role not in roles:
-            raise HTTPException(status_code=status.HTTP_403_FORBIDDEN, detail="Forbidden")
+            raise HTTPException(
+                status_code=status.HTTP_403_FORBIDDEN,
+                detail=api_error("forbidden", "Forbidden"),
+            )
         return user
 
     return role_dependency

backend/app/core/errors.py

@@ -0,0 +1,38 @@
+from __future__ import annotations
+
+from typing import Any
+
+
+def error_payload(code: str, message: str, details: Any, request_id: str) -> dict[str, Any]:
+    return {
+        "code": code,
+        "message": message,
+        "details": details,
+        "request_id": request_id,
+    }
+
+
+def api_error(code: str, message: str, details: Any = None) -> dict[str, Any]:
+    return {
+        "code": code,
+        "message": message,
+        "details": details,
+    }
+
+
+def http_status_to_code(status_code: int) -> str:
+    mapping = {
+        400: "bad_request",
+        401: "unauthorized",
+        403: "forbidden",
+        404: "not_found",
+        405: "method_not_allowed",
+        409: "conflict",
+        422: "validation_error",
+        429: "rate_limited",
+        500: "internal_error",
+        502: "bad_gateway",
+        503: "service_unavailable",
+        504: "gateway_timeout",
+    }
+    return mapping.get(status_code, f"http_{status_code}")

backend/app/core/security.py

@@ -1,5 +1,5 @@
 from datetime import datetime, timedelta, timezone
-from jose import jwt
+import jwt
 from passlib.context import CryptContext
 from app.core.config import get_settings
 

backend/app/main.py

@@ -1,12 +1,17 @@
 import asyncio
 import logging
+from uuid import uuid4
 from contextlib import asynccontextmanager
-from fastapi import FastAPI
+from fastapi import FastAPI, HTTPException, Request
+from fastapi.exceptions import RequestValidationError
 from fastapi.middleware.cors import CORSMiddleware
+from fastapi.responses import JSONResponse
+from starlette.exceptions import HTTPException as StarletteHTTPException
 from sqlalchemy import select
 from app.api.router import api_router
 from app.core.config import get_settings
 from app.core.db import SessionLocal
+from app.core.errors import error_payload, http_status_to_code
 from app.core.logging import configure_logging
 from app.core.security import hash_password
 from app.models.models import User
@@ -57,4 +62,67 @@ app.add_middleware(
     allow_methods=["*"],
     allow_headers=["*"],
 )
+
+
+@app.middleware("http")
+async def request_id_middleware(request: Request, call_next):
+    request_id = request.headers.get("x-request-id") or str(uuid4())
+    request.state.request_id = request_id
+    response = await call_next(request)
+    response.headers["X-Request-ID"] = request_id
+    return response
+
+
+@app.exception_handler(HTTPException)
+@app.exception_handler(StarletteHTTPException)
+async def http_exception_handler(request: Request, exc: HTTPException | StarletteHTTPException):
+    request_id = getattr(request.state, "request_id", str(uuid4()))
+    code = http_status_to_code(exc.status_code)
+    message = "Request failed"
+    details = None
+
+    if isinstance(exc.detail, str):
+        message = exc.detail
+    elif isinstance(exc.detail, dict):
+        code = str(exc.detail.get("code", code))
+        message = str(exc.detail.get("message", message))
+        details = exc.detail.get("details")
+    elif isinstance(exc.detail, list):
+        message = "Request validation failed"
+        details = exc.detail
+
+    return JSONResponse(
+        status_code=exc.status_code,
+        content=error_payload(code=code, message=message, details=details, request_id=request_id),
+    )
+
+
+@app.exception_handler(RequestValidationError)
+async def request_validation_exception_handler(request: Request, exc: RequestValidationError):
+    request_id = getattr(request.state, "request_id", str(uuid4()))
+    return JSONResponse(
+        status_code=422,
+        content=error_payload(
+            code="validation_error",
+            message="Request validation failed",
+            details=exc.errors(),
+            request_id=request_id,
+        ),
+    )
+
+
+@app.exception_handler(Exception)
+async def unhandled_exception_handler(request: Request, exc: Exception):
+    request_id = getattr(request.state, "request_id", str(uuid4()))
+    logger.exception("unhandled_exception request_id=%s", request_id, exc_info=exc)
+    return JSONResponse(
+        status_code=500,
+        content=error_payload(
+            code="internal_error",
+            message="Internal server error",
+            details=None,
+            request_id=request_id,
+        ),
+    )
+
 app.include_router(api_router, prefix=settings.api_v1_prefix)

backend/app/models/models.py

@@ -9,6 +9,8 @@ class User(Base):
 
     id: Mapped[int] = mapped_column(Integer, primary_key=True)
     email: Mapped[str] = mapped_column(String(255), unique=True, index=True, nullable=False)
+    first_name: Mapped[str | None] = mapped_column(String(120), nullable=True)
+    last_name: Mapped[str | None] = mapped_column(String(120), nullable=True)
     password_hash: Mapped[str] = mapped_column(String(255), nullable=False)
     role: Mapped[str] = mapped_column(String(20), nullable=False, default="viewer")
     created_at: Mapped[datetime] = mapped_column(DateTime(timezone=True), server_default=func.now(), nullable=False)

backend/app/schemas/user.py

@@ -5,6 +5,8 @@ from pydantic import BaseModel, EmailStr, field_validator
 class UserOut(BaseModel):
     id: int
     email: EmailStr
+    first_name: str | None = None
+    last_name: str | None = None
     role: str
     created_at: datetime
 
@@ -13,12 +15,16 @@ class UserOut(BaseModel):
 
 class UserCreate(BaseModel):
     email: EmailStr
+    first_name: str | None = None
+    last_name: str | None = None
     password: str
     role: str = "viewer"
 
 
 class UserUpdate(BaseModel):
     email: EmailStr | None = None
+    first_name: str | None = None
+    last_name: str | None = None
     password: str | None = None
     role: str | None = None
 

backend/app/services/alerts.py

@@ -11,6 +11,7 @@ from sqlalchemy import desc, func, select
 from sqlalchemy.ext.asyncio import AsyncSession
 
 from app.core.config import get_settings
+from app.core.errors import api_error
 from app.models.models import AlertDefinition, Metric, QueryStat, Target
 from app.schemas.alert import AlertStatusItem, AlertStatusResponse
 from app.services.collector import build_target_dsn
@@ -144,25 +145,40 @@ def get_standard_alert_reference() -> list[dict[str, str]]:
 
 def validate_alert_thresholds(comparison: str, warning_threshold: float | None, alert_threshold: float) -> None:
     if comparison not in _ALLOWED_COMPARISONS:
-        raise HTTPException(status_code=400, detail=f"Invalid comparison. Use one of {sorted(_ALLOWED_COMPARISONS)}")
+        raise HTTPException(
+            status_code=400,
+            detail=api_error(
+                "invalid_comparison",
+                f"Invalid comparison. Use one of {sorted(_ALLOWED_COMPARISONS)}",
+            ),
+        )
     if warning_threshold is None:
         return
 
     if comparison in {"gte", "gt"} and warning_threshold > alert_threshold:
-        raise HTTPException(status_code=400, detail="For gte/gt, warning_threshold must be <= alert_threshold")
+        raise HTTPException(
+            status_code=400,
+            detail=api_error("invalid_thresholds", "For gte/gt, warning_threshold must be <= alert_threshold"),
+        )
     if comparison in {"lte", "lt"} and warning_threshold < alert_threshold:
-        raise HTTPException(status_code=400, detail="For lte/lt, warning_threshold must be >= alert_threshold")
+        raise HTTPException(
+            status_code=400,
+            detail=api_error("invalid_thresholds", "For lte/lt, warning_threshold must be >= alert_threshold"),
+        )
 
 
 def validate_alert_sql(sql_text: str) -> str:
     sql = sql_text.strip().rstrip(";")
     lowered = sql.lower().strip()
     if not lowered.startswith("select"):
-        raise HTTPException(status_code=400, detail="Alert SQL must start with SELECT")
+        raise HTTPException(status_code=400, detail=api_error("invalid_alert_sql", "Alert SQL must start with SELECT"))
     if _FORBIDDEN_SQL_WORDS.search(lowered):
-        raise HTTPException(status_code=400, detail="Only read-only SELECT statements are allowed")
+        raise HTTPException(
+            status_code=400,
+            detail=api_error("invalid_alert_sql", "Only read-only SELECT statements are allowed"),
+        )
     if ";" in sql:
-        raise HTTPException(status_code=400, detail="Only a single SQL statement is allowed")
+        raise HTTPException(status_code=400, detail=api_error("invalid_alert_sql", "Only a single SQL statement is allowed"))
     return sql
 
 

backend/app/services/collector.py

@@ -1,6 +1,7 @@
 import asyncio
 import logging
 from datetime import datetime, timezone
+from random import uniform
 from sqlalchemy import select
 from sqlalchemy.ext.asyncio import AsyncSession
 from sqlalchemy.exc import SQLAlchemyError
@@ -15,6 +16,9 @@ logger = logging.getLogger(__name__)
 settings = get_settings()
 _failure_state: dict[int, dict[str, object]] = {}
 _failure_log_interval_seconds = 300
+_backoff_base_seconds = max(3, int(settings.poll_interval_seconds))
+_backoff_max_seconds = 300
+_backoff_jitter_factor = 0.15
 
 
 def build_target_dsn(target: Target) -> str:
@@ -181,31 +185,66 @@ async def collect_once() -> None:
     async with SessionLocal() as db:
         targets = (await db.scalars(select(Target))).all()
 
+    active_target_ids = {target.id for target in targets}
+    stale_target_ids = [target_id for target_id in _failure_state.keys() if target_id not in active_target_ids]
+    for stale_target_id in stale_target_ids:
+        _failure_state.pop(stale_target_id, None)
+
     for target in targets:
+        now = datetime.now(timezone.utc)
+        state = _failure_state.get(target.id)
+        if state:
+            next_attempt_at = state.get("next_attempt_at")
+            if isinstance(next_attempt_at, datetime) and now < next_attempt_at:
+                continue
+
         try:
             await collect_target(target)
             prev = _failure_state.pop(target.id, None)
             if prev:
+                first_failure_at = prev.get("first_failure_at")
+                downtime_seconds = None
+                if isinstance(first_failure_at, datetime):
+                    downtime_seconds = max(0, int((now - first_failure_at).total_seconds()))
                 logger.info(
-                    "collector_target_recovered target=%s after_failures=%s last_error=%s",
+                    "collector_target_recovered target=%s after_failures=%s downtime_seconds=%s last_error=%s",
                     target.id,
                     prev.get("count", 0),
+                    downtime_seconds,
                     prev.get("error"),
                 )
         except (OSError, SQLAlchemyError, asyncpg.PostgresError, Exception) as exc:
-            now = datetime.now(timezone.utc)
             current_error = str(exc)
+            error_class = exc.__class__.__name__
             state = _failure_state.get(target.id)
             if state is None:
+                next_delay = min(_backoff_max_seconds, _backoff_base_seconds)
+                jitter = next_delay * _backoff_jitter_factor
+                next_delay = max(1, int(next_delay + uniform(-jitter, jitter)))
+                next_attempt_at = now.timestamp() + next_delay
                 _failure_state[target.id] = {
                     "count": 1,
+                    "first_failure_at": now,
                     "last_log_at": now,
                     "error": current_error,
+                    "next_attempt_at": datetime.fromtimestamp(next_attempt_at, tz=timezone.utc),
                 }
-                logger.exception("collector_error target=%s err=%s", target.id, exc)
+                logger.warning(
+                    "collector_target_unreachable target=%s error_class=%s err=%s consecutive_failures=%s retry_in_seconds=%s",
+                    target.id,
+                    error_class,
+                    current_error,
+                    1,
+                    next_delay,
+                )
                 continue
 
             count = int(state.get("count", 0)) + 1
+            raw_backoff = min(_backoff_max_seconds, _backoff_base_seconds * (2 ** min(count - 1, 10)))
+            jitter = raw_backoff * _backoff_jitter_factor
+            next_delay = max(1, int(raw_backoff + uniform(-jitter, jitter)))
+            state["next_attempt_at"] = datetime.fromtimestamp(now.timestamp() + next_delay, tz=timezone.utc)
+
             last_log_at = state.get("last_log_at")
             last_logged_error = str(state.get("error", ""))
             should_log = False
@@ -220,18 +259,23 @@ async def collect_once() -> None:
             if should_log:
                 state["last_log_at"] = now
                 state["error"] = current_error
-                logger.error(
+                logger.warning(
-                    "collector_error_throttled target=%s err=%s consecutive_failures=%s",
+                    "collector_target_unreachable target=%s error_class=%s err=%s consecutive_failures=%s retry_in_seconds=%s",
                     target.id,
+                    error_class,
                     current_error,
                     count,
+                    next_delay,
                 )
 
 
 async def collector_loop(stop_event: asyncio.Event) -> None:
     while not stop_event.is_set():
+        cycle_started = asyncio.get_running_loop().time()
         await collect_once()
+        elapsed = asyncio.get_running_loop().time() - cycle_started
+        sleep_for = max(0.0, settings.poll_interval_seconds - elapsed)
         try:
-            await asyncio.wait_for(stop_event.wait(), timeout=settings.poll_interval_seconds)
+            await asyncio.wait_for(stop_event.wait(), timeout=sleep_for)
         except asyncio.TimeoutError:
             pass

backend/app/services/infra_probe.py

@@ -17,10 +17,10 @@ class DiskSpaceProvider:
 class NullDiskSpaceProvider(DiskSpaceProvider):
     async def get_free_bytes(self, target_host: str) -> DiskSpaceProbeResult:
         return DiskSpaceProbeResult(
-            source="none",
+            source="agentless",
             status="unavailable",
             free_bytes=None,
-            message=f"No infra probe configured for host {target_host}. Add SSH/Agent provider later.",
+            message=f"Agentless mode: host-level free disk is not available for {target_host}.",
         )
 
 

backend/requirements.txt

@@ -1,4 +1,5 @@
-fastapi==0.116.1
+fastapi==0.129.0
+starlette==0.52.1
 uvicorn[standard]==0.35.0
 gunicorn==23.0.0
 sqlalchemy[asyncio]==2.0.44
@@ -7,7 +8,7 @@ alembic==1.16.5
 pydantic==2.11.7
 pydantic-settings==2.11.0
 email-validator==2.2.0
-python-jose[cryptography]==3.5.0
+PyJWT==2.11.0
 passlib[argon2]==1.7.4
-cryptography==45.0.7
+cryptography==46.0.5
-python-multipart==0.0.20
+python-multipart==0.0.22

docker-compose.yml

@@ -18,8 +18,8 @@ services:
       retries: 10
 
   backend:
-    build:
+    image: nesterovicit/nexapg-backend:latest
-      context: ./backend
+    pull_policy: always
     container_name: nexapg-backend
     restart: unless-stopped
     environment:
@@ -47,10 +47,8 @@ services:
       - "${BACKEND_PORT}:8000"
 
   frontend:
-    build:
+    image: nesterovicit/nexapg-frontend:latest
-      context: ./frontend
+    pull_policy: always
-      args:
-        VITE_API_URL: ${VITE_API_URL}
     container_name: nexapg-frontend
     restart: unless-stopped
     depends_on:

frontend/Dockerfile

@@ -7,8 +7,9 @@ ARG VITE_API_URL=/api/v1
 ENV VITE_API_URL=${VITE_API_URL}
 RUN npm run build
 
-FROM nginx:1.29-alpine
+FROM nginx:1.29-alpine-slim
+RUN apk upgrade --no-cache
 COPY nginx.conf /etc/nginx/conf.d/default.conf
 COPY --from=build /app/dist /usr/share/nginx/html
 EXPOSE 80
-HEALTHCHECK --interval=30s --timeout=3s --retries=5 CMD wget -qO- http://127.0.0.1/ || exit 1
+HEALTHCHECK --interval=30s --timeout=3s --retries=5 CMD nginx -t || exit 1

frontend/src/App.jsx

@@ -22,6 +22,7 @@ function Layout({ children }) {
   const { me, logout, uiMode, setUiMode, alertToasts, dismissAlertToast, serviceUpdateAvailable } = useAuth();
   const navigate = useNavigate();
   const navClass = ({ isActive }) => `nav-btn${isActive ? " active" : ""}`;
+  const fullName = [me?.first_name, me?.last_name].filter(Boolean).join(" ").trim();
 
   return (
     <div className="shell">
@@ -101,8 +102,9 @@ function Layout({ children }) {
             </button>
             <small>{uiMode === "easy" ? "Simple health guidance" : "Advanced DBA metrics"}</small>
           </div>
-          <div>{me?.email}</div>
+          <div className="profile-name">{fullName || me?.email}</div>
-          <div className="role">{me?.role}</div>
+          {fullName && <div className="profile-email">{me?.email}</div>}
+          <div className="role profile-role">{me?.role}</div>
           <NavLink to="/user-settings" className={({ isActive }) => `profile-btn${isActive ? " active" : ""}`}>
             User Settings
           </NavLink>

frontend/src/api.js

@@ -35,8 +35,21 @@ export async function apiFetch(path, options = {}, tokens, onUnauthorized) {
     }
   }
   if (!res.ok) {
-    const txt = await res.text();
+    const raw = await res.text();
-    throw new Error(txt || `HTTP ${res.status}`);
+    let parsed = null;
+    try {
+      parsed = raw ? JSON.parse(raw) : null;
+    } catch {
+      parsed = null;
+    }
+
+    const message = parsed?.message || raw || `HTTP ${res.status}`;
+    const err = new Error(message);
+    err.status = res.status;
+    err.code = parsed?.code || null;
+    err.details = parsed?.details || null;
+    err.requestId = parsed?.request_id || res.headers.get("x-request-id") || null;
+    throw err;
   }
   if (res.status === 204) return null;
   return res.json();

frontend/src/pages/AdminUsersPage.jsx

@@ -19,8 +19,11 @@ const TEMPLATE_VARIABLES = [
 
 export function AdminUsersPage() {
   const { tokens, refresh, me } = useAuth();
+  const emptyCreateForm = { email: "", first_name: "", last_name: "", password: "", role: "viewer" };
   const [users, setUsers] = useState([]);
-  const [form, setForm] = useState({ email: "", password: "", role: "viewer" });
+  const [form, setForm] = useState(emptyCreateForm);
+  const [editingUserId, setEditingUserId] = useState(null);
+  const [editForm, setEditForm] = useState({ email: "", first_name: "", last_name: "", password: "", role: "viewer" });
   const [emailSettings, setEmailSettings] = useState({
     enabled: false,
     smtp_host: "",
@@ -79,7 +82,7 @@ export function AdminUsersPage() {
     e.preventDefault();
     try {
       await apiFetch("/admin/users", { method: "POST", body: JSON.stringify(form) }, tokens, refresh);
-      setForm({ email: "", password: "", role: "viewer" });
+      setForm(emptyCreateForm);
       await load();
     } catch (e) {
       setError(String(e.message || e));
@@ -95,6 +98,39 @@ export function AdminUsersPage() {
     }
   };
 
+  const startEdit = (user) => {
+    setEditingUserId(user.id);
+    setEditForm({
+      email: user.email || "",
+      first_name: user.first_name || "",
+      last_name: user.last_name || "",
+      password: "",
+      role: user.role || "viewer",
+    });
+  };
+
+  const cancelEdit = () => {
+    setEditingUserId(null);
+    setEditForm({ email: "", first_name: "", last_name: "", password: "", role: "viewer" });
+  };
+
+  const saveEdit = async (userId) => {
+    try {
+      const payload = {
+        email: editForm.email,
+        first_name: editForm.first_name.trim() || null,
+        last_name: editForm.last_name.trim() || null,
+        role: editForm.role,
+      };
+      if (editForm.password.trim()) payload.password = editForm.password;
+      await apiFetch(`/admin/users/${userId}`, { method: "PUT", body: JSON.stringify(payload) }, tokens, refresh);
+      cancelEdit();
+      await load();
+    } catch (e) {
+      setError(String(e.message || e));
+    }
+  };
+
   const saveSmtp = async (e) => {
     e.preventDefault();
     setError("");
@@ -165,6 +201,22 @@ export function AdminUsersPage() {
           <p className="muted">Create accounts and manage access roles.</p>
         </div>
         <form className="grid three admin-user-form" onSubmit={create}>
+          <div className="admin-field">
+            <label>First Name</label>
+            <input
+              value={form.first_name}
+              placeholder="Jane"
+              onChange={(e) => setForm({ ...form, first_name: e.target.value })}
+            />
+          </div>
+          <div className="admin-field">
+            <label>Last Name</label>
+            <input
+              value={form.last_name}
+              placeholder="Doe"
+              onChange={(e) => setForm({ ...form, last_name: e.target.value })}
+            />
+          </div>
           <div className="admin-field">
             <label>Email</label>
             <input value={form.email} placeholder="user@example.com" onChange={(e) => setForm({ ...form, email: e.target.value })} />
@@ -197,6 +249,7 @@ export function AdminUsersPage() {
           <thead>
             <tr>
               <th>ID</th>
+              <th>Name</th>
               <th>Email</th>
               <th>Role</th>
               <th>Action</th>
@@ -206,11 +259,70 @@ export function AdminUsersPage() {
             {users.map((u) => (
               <tr key={u.id} className="admin-user-row">
                 <td className="user-col-id">{u.id}</td>
-                <td className="user-col-email">{u.email}</td>
+                <td className="user-col-name">
-                <td>
+                  {editingUserId === u.id ? (
-                  <span className={`pill role-pill role-${u.role}`}>{u.role}</span>
+                    <div className="admin-inline-grid two">
+                      <input
+                        value={editForm.first_name}
+                        placeholder="First name"
+                        onChange={(e) => setEditForm({ ...editForm, first_name: e.target.value })}
+                      />
+                      <input
+                        value={editForm.last_name}
+                        placeholder="Last name"
+                        onChange={(e) => setEditForm({ ...editForm, last_name: e.target.value })}
+                      />
+                    </div>
+                  ) : (
+                    <span className="user-col-name-value">{[u.first_name, u.last_name].filter(Boolean).join(" ") || "-"}</span>
+                  )}
+                </td>
+                <td className="user-col-email">
+                  {editingUserId === u.id ? (
+                    <input
+                      value={editForm.email}
+                      placeholder="user@example.com"
+                      onChange={(e) => setEditForm({ ...editForm, email: e.target.value })}
+                    />
+                  ) : (
+                    u.email
+                  )}
                 </td>
                 <td>
+                  {editingUserId === u.id ? (
+                    <select value={editForm.role} onChange={(e) => setEditForm({ ...editForm, role: e.target.value })}>
+                      <option value="viewer">viewer</option>
+                      <option value="operator">operator</option>
+                      <option value="admin">admin</option>
+                    </select>
+                  ) : (
+                    <span className={`pill role-pill role-${u.role}`}>{u.role}</span>
+                  )}
+                </td>
+                <td className="admin-user-actions">
+                  {editingUserId === u.id && (
+                    <input
+                      type="password"
+                      className="admin-inline-password"
+                      value={editForm.password}
+                      placeholder="New password (optional)"
+                      onChange={(e) => setEditForm({ ...editForm, password: e.target.value })}
+                    />
+                  )}
+                  {editingUserId === u.id ? (
+                    <>
+                      <button className="table-action-btn primary small-btn" onClick={() => saveEdit(u.id)}>
+                        Save
+                      </button>
+                      <button className="table-action-btn small-btn" onClick={cancelEdit}>
+                        Cancel
+                      </button>
+                    </>
+                  ) : (
+                    <button className="table-action-btn edit small-btn" onClick={() => startEdit(u)}>
+                      Edit
+                    </button>
+                  )}
                   {u.id !== me.id && (
                     <button className="table-action-btn delete small-btn" onClick={() => remove(u.id)}>
                       <span aria-hidden="true">

frontend/src/pages/QueryInsightsPage.jsx

@@ -1,4 +1,4 @@
-import React, { useEffect, useState } from "react";
+import React, { useEffect, useRef, useState } from "react";
 import { apiFetch } from "../api";
 import { useAuth } from "../state";
 
@@ -62,6 +62,7 @@ function buildQueryTips(row) {
 
 export function QueryInsightsPage() {
   const { tokens, refresh } = useAuth();
+  const refreshRef = useRef(refresh);
   const [targets, setTargets] = useState([]);
   const [targetId, setTargetId] = useState("");
   const [rows, setRows] = useState([]);
@@ -71,6 +72,10 @@ export function QueryInsightsPage() {
   const [error, setError] = useState("");
   const [loading, setLoading] = useState(true);
 
+  useEffect(() => {
+    refreshRef.current = refresh;
+  }, [refresh]);
+
   useEffect(() => {
     (async () => {
       try {
@@ -89,17 +94,26 @@ export function QueryInsightsPage() {
 
   useEffect(() => {
     if (!targetId) return;
+    let active = true;
     (async () => {
       try {
-        const data = await apiFetch(`/targets/${targetId}/top-queries`, {}, tokens, refresh);
+        const data = await apiFetch(`/targets/${targetId}/top-queries`, {}, tokens, refreshRef.current);
+        if (!active) return;
         setRows(data);
-        setSelectedQuery(data[0] || null);
+        setSelectedQuery((prev) => {
-        setPage(1);
+          if (!prev) return data[0] || null;
+          const keep = data.find((row) => row.queryid === prev.queryid);
+          return keep || data[0] || null;
+        });
+        setPage((prev) => (prev === 1 ? prev : 1));
       } catch (e) {
-        setError(String(e.message || e));
+        if (active) setError(String(e.message || e));
       }
     })();
-  }, [targetId, tokens, refresh]);
+    return () => {
+      active = false;
+    };
+  }, [targetId, tokens?.accessToken, tokens?.refreshToken]);
 
   const dedupedByQueryId = [...rows].reduce((acc, row) => {
     if (!row?.queryid) return acc;

frontend/src/pages/TargetDetailPage.jsx

@@ -41,6 +41,19 @@ function formatNumber(value, digits = 2) {
   return Number(value).toFixed(digits);
 }
 
+function formatHostMetricUnavailable() {
+  return "N/A (agentless)";
+}
+
+function formatDiskSpaceAgentless(diskSpace) {
+  if (!diskSpace) return formatHostMetricUnavailable();
+  if (diskSpace.free_bytes !== null && diskSpace.free_bytes !== undefined) {
+    return formatBytes(diskSpace.free_bytes);
+  }
+  if (diskSpace.status === "unavailable") return formatHostMetricUnavailable();
+  return "-";
+}
+
 function MetricsTooltip({ active, payload, label }) {
   if (!active || !payload || payload.length === 0) return null;
   const row = payload[0]?.payload || {};
@@ -63,6 +76,10 @@ function didMetricSeriesChange(prev = [], next = []) {
   return prevLast?.ts !== nextLast?.ts || Number(prevLast?.value) !== Number(nextLast?.value);
 }
 
+function isTargetUnreachableError(err) {
+  return err?.code === "target_unreachable" || err?.status === 503;
+}
+
 async function loadMetric(targetId, metric, range, tokens, refresh) {
   const { from, to } = toQueryRange(range);
   return apiFetch(
@@ -86,6 +103,7 @@ export function TargetDetailPage() {
   const [targetMeta, setTargetMeta] = useState(null);
   const [owners, setOwners] = useState([]);
   const [groupTargets, setGroupTargets] = useState([]);
+  const [offlineState, setOfflineState] = useState(null);
   const [error, setError] = useState("");
   const [loading, setLoading] = useState(true);
   const refreshRef = useRef(refresh);
@@ -101,22 +119,16 @@ export function TargetDetailPage() {
         setLoading(true);
       }
       try {
-        const [connections, xacts, cache, locksTable, activityTable, overviewData, targetInfo, ownerRows, allTargets] = await Promise.all([
+        const [connections, xacts, cache, targetInfo, ownerRows, allTargets] = await Promise.all([
           loadMetric(id, "connections_total", range, tokens, refreshRef.current),
           loadMetric(id, "xacts_total", range, tokens, refreshRef.current),
           loadMetric(id, "cache_hit_ratio", range, tokens, refreshRef.current),
-          apiFetch(`/targets/${id}/locks`, {}, tokens, refreshRef.current),
-          apiFetch(`/targets/${id}/activity`, {}, tokens, refreshRef.current),
-          apiFetch(`/targets/${id}/overview`, {}, tokens, refreshRef.current),
           apiFetch(`/targets/${id}`, {}, tokens, refreshRef.current),
           apiFetch(`/targets/${id}/owners`, {}, tokens, refreshRef.current),
           apiFetch("/targets", {}, tokens, refreshRef.current),
         ]);
         if (!active) return;
         setSeries({ connections, xacts, cache });
-        setLocks(locksTable);
-        setActivity(activityTable);
-        setOverview(overviewData);
         setTargetMeta(targetInfo);
         setOwners(ownerRows);
         const groupId = targetInfo?.tags?.monitor_group_id;
@@ -128,6 +140,34 @@ export function TargetDetailPage() {
         } else {
           setGroupTargets([]);
         }
+        try {
+          const [locksTable, activityTable, overviewData] = await Promise.all([
+            apiFetch(`/targets/${id}/locks`, {}, tokens, refreshRef.current),
+            apiFetch(`/targets/${id}/activity`, {}, tokens, refreshRef.current),
+            apiFetch(`/targets/${id}/overview`, {}, tokens, refreshRef.current),
+          ]);
+          if (!active) return;
+          setLocks(locksTable);
+          setActivity(activityTable);
+          setOverview(overviewData);
+          setOfflineState(null);
+        } catch (liveErr) {
+          if (!active) return;
+          if (isTargetUnreachableError(liveErr)) {
+            setLocks([]);
+            setActivity([]);
+            setOverview(null);
+            setOfflineState({
+              message:
+                "Target is currently unreachable. Check host/port, network route, SSL mode, and database availability.",
+              host: liveErr?.details?.host || targetInfo?.host || "-",
+              port: liveErr?.details?.port || targetInfo?.port || "-",
+              requestId: liveErr?.requestId || null,
+            });
+          } else {
+            throw liveErr;
+          }
+        }
         setError("");
       } catch (e) {
         if (active) setError(String(e.message || e));
@@ -268,6 +308,17 @@ export function TargetDetailPage() {
         <span className="muted">Responsible users:</span>
         {owners.length > 0 ? owners.map((item) => <span key={item.user_id} className="owner-pill">{item.email}</span>) : <span className="muted">none assigned</span>}
       </div>
+      {offlineState && (
+        <div className="card target-offline-card">
+          <h3>Target Offline</h3>
+          <p>{offlineState.message}</p>
+          <div className="target-offline-meta">
+            <span><strong>Host:</strong> {offlineState.host}</span>
+            <span><strong>Port:</strong> {offlineState.port}</span>
+            {offlineState.requestId ? <span><strong>Request ID:</strong> {offlineState.requestId}</span> : null}
+          </div>
+        </div>
+      )}
       {uiMode === "easy" && overview && easySummary && (
         <>
           <div className={`card easy-status ${easySummary.health}`}>
@@ -346,6 +397,9 @@ export function TargetDetailPage() {
       {uiMode === "dba" && overview && (
         <div className="card">
           <h3>Database Overview</h3>
+          <p className="muted" style={{ marginTop: 2 }}>
+            Agentless mode: host-level CPU, RAM, and free-disk metrics are not available.
+          </p>
           <div className="grid three overview-kv">
             <div><span>PostgreSQL Version</span><strong>{overview.instance.server_version || "-"}</strong></div>
             <div>
@@ -366,8 +420,8 @@ export function TargetDetailPage() {
             <div title="Total WAL directory size (when available)">
               <span>WAL Size</span><strong>{formatBytes(overview.storage.wal_directory_size_bytes)}</strong>
             </div>
-            <div title="Optional metric via future Agent/SSH provider">
+            <div title={overview.storage.disk_space?.message || "Agentless mode: host-level free disk is unavailable."}>
-              <span>Free Disk</span><strong>{formatBytes(overview.storage.disk_space.free_bytes)}</strong>
+              <span>Free Disk</span><strong>{formatDiskSpaceAgentless(overview.storage.disk_space)}</strong>
             </div>
             <div title="Replication replay delay on standby">
               <span>Replay Lag</span>
@@ -378,6 +432,12 @@ export function TargetDetailPage() {
             <div><span>Replication Slots</span><strong>{overview.replication.replication_slots_count ?? "-"}</strong></div>
             <div><span>Repl Clients</span><strong>{overview.replication.active_replication_clients ?? "-"}</strong></div>
             <div><span>Autovacuum Workers</span><strong>{overview.performance.autovacuum_workers ?? "-"}</strong></div>
+            <div title="Host CPU requires OS-level telemetry">
+              <span>Host CPU</span><strong>{formatHostMetricUnavailable()}</strong>
+            </div>
+            <div title="Host RAM requires OS-level telemetry">
+              <span>Host RAM</span><strong>{formatHostMetricUnavailable()}</strong>
+            </div>
           </div>
 
           <div className="grid two">

frontend/src/styles.css

@@ -195,6 +195,23 @@ a {
   color: #d7e4fa;
 }
 
+.profile-name {
+  font-size: 15px;
+  font-weight: 700;
+  line-height: 1.25;
+}
+
+.profile-email {
+  margin-top: 2px;
+  font-size: 12px;
+  color: #a6bcda;
+  word-break: break-all;
+}
+
+.profile-role {
+  margin-top: 4px;
+}
+
 .mode-switch-block {
   margin-bottom: 12px;
   padding: 10px;
@@ -1255,6 +1272,31 @@ td {
   font-weight: 600;
 }
 
+.user-col-name-value {
+  font-weight: 600;
+}
+
+.admin-inline-grid {
+  display: grid;
+  gap: 8px;
+}
+
+.admin-inline-grid.two {
+  grid-template-columns: repeat(2, minmax(0, 1fr));
+}
+
+.admin-inline-password {
+  min-width: 190px;
+}
+
+.admin-user-actions {
+  display: flex;
+  gap: 8px;
+  align-items: center;
+  justify-content: flex-start;
+  flex-wrap: wrap;
+}
+
 .role-pill {
   display: inline-flex;
   align-items: center;
@@ -1980,6 +2022,29 @@ select:-webkit-autofill {
   font-size: 12px;
 }
 
+.target-offline-card {
+  border-color: #a85757;
+  background: linear-gradient(130deg, #2c1724 0%, #1f1f38 100%);
+}
+
+.target-offline-card h3 {
+  margin: 0 0 8px;
+  color: #fecaca;
+}
+
+.target-offline-card p {
+  margin: 0 0 10px;
+  color: #fde2e2;
+}
+
+.target-offline-meta {
+  display: flex;
+  flex-wrap: wrap;
+  gap: 16px;
+  font-size: 12px;
+  color: #d4d4f5;
+}
+
 .chart-tooltip {
   background: #0f1934ee;
   border: 1px solid #2f4a8b;

ops/scripts/bootstrap-compose.sh

@@ -0,0 +1,41 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+# Usage:
+#   bash bootstrap-compose.sh
+#   BASE_URL="https://git.nesterovic.cc/nessi/NexaPG/raw/branch/main" bash bootstrap-compose.sh
+
+BASE_URL="${BASE_URL:-https://git.nesterovic.cc/nessi/NexaPG/raw/branch/main}"
+
+echo "[bootstrap] Using base URL: ${BASE_URL}"
+
+fetch_file() {
+  local path="$1"
+  local out="$2"
+
+  if command -v wget >/dev/null 2>&1; then
+    wget -q -O "${out}" "${BASE_URL}/${path}"
+  elif command -v curl >/dev/null 2>&1; then
+    curl -fsSL "${BASE_URL}/${path}" -o "${out}"
+  else
+    echo "[bootstrap] ERROR: wget or curl is required"
+    exit 1
+  fi
+}
+
+fetch_file "docker-compose.yml" "docker-compose.yml"
+fetch_file ".env.example" ".env.example"
+fetch_file "Makefile" "Makefile"
+
+if [[ ! -f ".env" ]]; then
+  cp .env.example .env
+  echo "[bootstrap] Created .env from .env.example"
+else
+  echo "[bootstrap] .env already exists, keeping it"
+fi
+
+echo
+echo "[bootstrap] Next steps:"
+echo "  1) Edit .env (set JWT_SECRET_KEY and ENCRYPTION_KEY at minimum)"
+echo "  2) Run: make up"
+echo