Security Management and Architecture
- security
- Architecture
- Systems
- Disaster Recovery
- Risk assessment
- Curtis
In this guide we will cover:
- FAQs around Security Management and Architecture
How is customer backup data retained with Halo?
Transactional backups are taken hourly, retained for 3 days.
Daily full backups are retained for 2 weeks.
Monthly full backups are retained for 90 days.
UAT environments only have the daily backups taken, also retained for 2 weeks.
Where is customer data stored?
Data is stored for the customer on their relevant AWS hosted data centre zone within a regional S3 bucket solution
How is customer data encrypted at rest?
All databases and backups are encrypted at rest using AES-256 key encryption
What IPs does Halo use for X component/process?
IPs relating to Halo internal components will not be useful to an external party, they are held within a private subnet(making them inaccessible to the outside internet) and change regularly.
Is Halo compliant with X standard/guideline?
All certifications and security reports completed by Halo can be viewed on our trust centre portal - https://trust.usehalo.com/
What is the Halo release schedule/procedure for patches and upgrades?
We have two different schedule paths available for customers, stable and beta. For clarity, upgrades refer to a change to the second number of a version IE 2.131.10 to 2.132.8 and may include changes to the schema, and a patch is a change to the last number IE 2.132.8 to 2.132.9 and will not include schema changes, only bug fixes/improvements.
The beta path is subject to version upgrades approximately every two weeks, and is rolled out to customers over two days. This will be out of working hours for all regions. Version rollouts may be delayed for the beta path during release candidate periods, as the development team prioritises bug fixes and optimisations for the upcoming stable.
The stable path is subject to version upgrades quarterly, and is rolled out to customers over a course of several weeks. The version that will become the stable (release candidate) is announced and available to customers to opt into 4 weeks before release as the stable. The upgrade schedule is communicated to customers 2 weeks before the start of the upgrade process. This will be out of working hours for all regions.
Patches are rolled out to customers at the discretion of the Halo team, this may include during working hours.
What is the target availability for Halo instances?
We target a 99.95% uptime for all production instances
What are Halos recovery targets during an incident?
Halo sets a recovery time objective of 4 hours for any given incident. This is the time taken during a disaster to restore a base level of operation. We set a recovery point target of 1 hour, meaning that data within the instance will be preserved from within an hour of the disaster incident occurring.
How are different customers separated within Halo's system architecture?
Customers are separated only at the database level. Underlying backend infrastructure and application level systems are shared between a subset of customers, but each customer has their own database