Skip to content

Latest commit

 

History

History
145 lines (117 loc) · 5.9 KB

data-leakage-backup-and-recovery.md

File metadata and controls

145 lines (117 loc) · 5.9 KB
Raw

Data leakage, backup and recovery

Data leakage

  • Any sort of unauthorized disclosure of sensitive information from anyone/any system.
  • Includes emails, malicious links, device theft etc.
  • Data leakage leads to
    • loss of trust e.g. trust to governments decreased during late years
    • loss of profit e.g. Sony lost profit of their movies after they were leaked before publishing

Data leakage threats

External threats

  • Corporate espionage, phishing, malware
  • Business partners, consultants when company outsources
    • Less surveillance than own employees.

Internal threats

  • Also known as insider threats
  • Dangers are greater than external threats as they'll have greater access to the company
  • See also insider attacks
  • E.g. eavesdropping, shoulder surfing, dumpster diving can be used to acquire data.

Data loss prevention

  • Also known as DLP
  • Identification and monitoring of important information that is not to be shared outside the organization.
  • Can block, notify the sender or lets admins to analyze, react and report to sensitive data in transit.
  • Important tool for enterprise message systems
  • Uses different techniques of data access control
    • E.g. when e-mailing where content is scanned for fingerprints, classifications and bank account numbers.

Data backup

  • Process of making a duplicate copy of data that already exists.
  • Protects against data loss and corruption as it can lead to great financial damages.
  • No backup = Far more suspectable to all sorts of attacks, especially ransomware.

Backup mediums

Magnetic tapes

  • Oldest form, still used by many enterprises.
  • Retention time: ≈30 years
  • 📝 To pull anything off the tape, you have to fast-forward to wherever the correct file is stored
    • Good for restoring systems in one go.
    • Bad for incremental backups or storing a few files.
    • ❗ Only way to tell if backups are working is to fully restore from the tape and check if it's correctly restored.

Optical disks

  • 2 to 3 times slower than hard drives
  • Retention time: ≈25 years

Hard disks

  • Cheaper, easily accessible
  • Less stability than magnetic tapes
  • Retention time: ≈9-20 years

SSD disks

  • Includes also usb drives known as Flash storage or thumb-drive.
  • Resistant to shock, temperature, being run through the washing machine
  • Retention time: ≈10 years

Cloud storage

  • Requires little infrastructure
  • Depends on stable internet connection
  • No retention time, high reliability

SD and micro-SD

  • Little capacity and pricy.
  • Retention time: ≈10 years

Steps of data backup strategy

  1. Identify important data
    • because backing-up everything is too costly and takes up much storage.
  2. Choose appropriate backup media
    • Reliable, solid, preferably cheap
    • E.g. USBs or portable media for personal users, and HDD/SDDs for companies for more speed.
  3. Choose the appropriate backup strategy
    • Check features such as scheduling, monitoring file changes to update back-ups, protocols, integrations...
    • Paid vs Free
      • Free requires more knowledge and work, training costs (one time)
        • E.g. in Linux, set cron job from point A to B
      • Paid versions has recurring license costs including training
  4. Choose appropriate RAID levels
    • RAID 1
      • 2 disks
      • All that are written in disk A is also written to B, if one disk fails, other works
    • RAID 5
      • 3 disks
      • If A fails you can reconstruct based on data in B and C
    • RAIDing is not only for backups, can also use for faster read and writes
      • E.g. BIG = Everything is seen as one drive. File is written two all of them. Crazy write & read speeds. If single disk dies all data is gone.
  5. Choose the appropriate backup method
    • Cold backup
      • Performed while system is not in use.
      • E.g. at nights, during weekends.
    • Hot backup
      • Performed when system is still used.
      • E.g. you type a document, power shortage happens but it's still saved.
    • Warm backup
      • Hybrid of the two.
      • Period backups while system is in use, but you might lose longer time period than hot backup.
  6. Choose the appropriate backup locations
    • Three options:
      1. On-site: Same building / room
        • Susceptible to same types of problems like other servers.
        • If there's a breach, fire or earthquake = all data are gone
      2. Off-site: backup is performed at a remote location
      3. Cloud:
        • Most secure: most cost effective and safe where data won't be loss, no electricity, no hardware, no maintainable.
        • Can be replicated in same building, different buildings in same data center or different regions.
        • Can have privacy/trust issues: encrypt
  7. Choose the backup type
    • Full backup: Costly, you back up everything
    • Incremental backup
      • Backs-up on each change of the previous back-up
      • When restoring, you need to restore everything from the first full back-up
    • Differential backup:
      • Back-ups on difference to the initial backup on each backup.s
      • Faster restoring time as you only need the last point and the initial full back-up
  8. Appropriate backup solution: Combination of all of this
  9. Perform a recovery test
    • Ensure you can recover a data that's lost with DR tests e.g. twice a year.
    • Recovery drill
      • Simulating data tier outage
      • Recovering
      • Validate application integrity post recovery

Data recovery

  • Recovering lost data
  • Reasons
    • Accidental lost e.g. • natural disaster • corrupted data
    • Or can be intentionally destroyed
  • DR stands for "Disaster Recovery"
  • Most of data is recoverable but you can have recovery failure if backed up data becomes corrupt.