Azure Files NFS

Required Roles for VM

Role	Scope	When to Use
Storage File Data Privileged Contributor	Storage Account	Read/write access to NFS share
Storage File Data Privileged Reader	Storage Account	Read-only access
Storage Account Contributor	Storage Account	If VM needs to manage the storage account itself

Enable Managed Identity on VM

# System-assigned
az vm identity assign \
  --resource-group <rg> \
  --name <vm-name>

Assign RBAC Role via Azure CLI

# Get VM's managed identity principal ID
PRINCIPAL_ID=$(az vm show \
  --resource-group <rg> \
  --name <vm-name> \
  --query identity.principalId -o tsv)

# Get Storage Account resource ID
SA_ID=$(az storage account show \
  --resource-group <rg> \
  --name <storage-account> \
  --query id -o tsv)

# Assign role
az role assignment create \
  --assignee $PRINCIPAL_ID \
  --role "Storage File Data Privileged Contributor" \
  --scope $SA_ID

Prerequisites Checklist

Requirement	Detail
Storage tier	Premium FileStorage only (NFS not on standard)
Hierarchical namespace	Not required (unlike Blob NFS)
Secure transfer	Must be disabled (NFS doesn’t support HTTPS)
Network	Private endpoint or VNet service endpoint required
Protocol	NFSv4.1 only
OS	Linux VM only

Mount on the VM

# Install NFS client
sudo apt-get install nfs-common   # Ubuntu/Debian
sudo yum install nfs-utils        # RHEL/CentOS

# Mount
sudo mount -t nfs \
  <storage-account>.file.core.windows.net:/<storage-account>/<share-name> \
  /mount/point \
  -o vers=4,minorversion=1,sec=sys

Key Points

• No Kerberos — Azure Files NFS uses sec=sys (UID/GID based), not identity-based auth
• RBAC controls management plane access; actual file permissions are POSIX/chmod on the share
• No SMB/Windows — NFS shares can’t be accessed via SMB simultaneously
• Private endpoint is strongly recommended over service endpoint for production

Task:

Follow above steps and create 2 vm and mount azurefileshare, upload files and check in 2nd vm

Databases Overview

A database offers a structured approach for:

• Storing and retrieving data
• User management
• Connection management

Types of Databases

• Relational Databases
• NoSQL Databases
  • Key-Value Stores
  • Document Databases
  • Column Databases
  • Graph Databases
• Blockchain Ledger
• Cache Servers
• Data Warehouses

Relational Databases

• Data is organized in tables with rows and columns
• Each row represents a record
• Tables have relations between them
• Interactions use a formal language based on standard SQL

Examples

Database	Notes
SQL Server	Microsoft enterprise RDBMS
Oracle	Large enterprise, high-volume
PostgreSQL	Open-source, extensible
MySQL	Web apps, high read workloads
DB2	IBM enterprise database

Self-Hosted (On-Premise) Setup

Server Side

1. Install the Database Engine (Server side of DBMS)
2. Configure users
3. Any additional configurations required

Client Side

• Install database client software for users
• Install database client libraries for applications

Ongoing Administrative Activities

Activity	Description
User Management	Adding and managing users
Backup	Regular database backups
Replication	Read-only and write-supported replicas
Performance Tuning	Query and index optimization
Patching	Applying security and version updates

On-premise: You are responsible for ALL of the above activities.

Database as a Service (Cloud)

Public cloud platforms offer managed databases where administrative overhead is dramatically reduced:

Activity	Self-Hosted	Cloud (DBaaS)
Backup	Manual setup	Single click
Replication	Manual config	Single click
Patching	Manual	Automated (set time window)
Performance Tuning	Fully manual	Partially automated

Cloud Database Offerings

AWS

• MySQL (MariaDB)
• PostgreSQL
• SQL Server
• Oracle
• DB2
• Aurora (MySQL and PostgreSQL compatible — AWS native)

Azure

• Microsoft SQL Server
• MySQL (MariaDB)
• PostgreSQL

Practical Difference: Self-Hosted vs DBaaS

Self-Hosted MySQL (on Linux VM)

• Install the database manually
• Create users manually
• Grant permissions manually
• All administrative activities remain your responsibility

MySQL as a Service (Cloud)

• Set parameters → database is up and ready
• Connect and start using immediately
• Most administrative activities available as easy one-click operations