v4.1.2

In addition to minor enhancements and fixes, this maintenance release contains the following

• Templates now include option for AS3 declaration URL
  F5 CFTs now include the option of including the URL of an AS3 declaration. This allows you to specify the BIG-IP configuration you want on your newly created BIG-IP VE(s). In templates such as autoscale, where an F5-recommended configuration is deployed by default, specifying an AS3 declaration URL will override the default configuration with your declaration. For information on AS3 and composing declarations, see https://clouddocs.f5.com/products/extensions/f5-appsvcs-extension/3.5.1/)

Bug fixes in this release

• /var/ directory full on autoscale solutions (cloud-libs)
• Autoscale should perform retries on authentication (cloud-libs)
• Modify autoscale.js to check to see if --cluster update is already running (cloud-libs)
• Install ilx package option should not be saved in reboot script (cloud-libs)
• Setting ssl-port in cloud-libs onboard responds with unknown error code (cloud-libs)
• Autoscale NLB AS3 declaration is incorrect
• Options not fully populated in run_autoscale_backup.sh
• Failover custom-config.sh fails due to lack of eth2 interface
• Instance type update to support additional instance types

Released on 2/7/19

v4.1.1

This maintenance release contains the following change:

• In the autoscale via-nlb templates (currently experimental), resolved an issue that could cause deployment of the default application service configuration to fail.

Released on 1/8/19

v4.1.0

In addition to minor enhancements and fixes, this release contains the following:

Supported

• Supported templates now install AS3
  All of the supported templates for AWS now install F5’s Application Services Extension 3 (AS3) version 3.5.1 (LTS) on the BIG-IP VEs by default. AS3 uses a declarative configuration model, meaning you send a declaration file using a single Rest API call. See the AS3 documentation (https://clouddocs.f5.com/products/extensions/f5-appsvcs-extension/3.5.1/) for details on how to use AS3 on your BIG-IP VE(s).

Experimental

In addition to minor enhancements and fixes, this release contains the following:

• New Quick Start template
  This release contains the initial release of our quick start templates with a quick start for AWS. The quick start template allows you to quickly launch an auto scale group of BIG-IP VEs into AWS to demonstrate a typical micro-services use case and/or Blue and Green deployment strategy. This template deploys a full demo application stack, including the Network, a Bastion Host, and an auto scale group of BIG-IP VEs in front of two application Auto Scale Groups.

Released on 12/13/18

v4.0.0

Supported templates in 4.0.0

In addition to minor enhancements and fixes, this maintenance release contains the following:

• New utility licenses for PAYG templates
  You can now choose from the following new options for many of the PAYG (pay as you go) license: Per App VE, Advanced WAF, Per App VE Advanced WAF. See the matrix (https://github.com/F5Networks/f5-aws-cloudformation/blob/master/aws-bigip-version-matrix.md) for details and applicable templates. See https://www.f5.com/pdf/products/big-ip-per-app-virtual-edition-overview.pdf for more information on Per App VE.

• Support for Static IP addresses
  Previously, AWS templates leveraged DHCP to assign addresses to its interfaces. This facilitates automation scenarios where the source of truth for your deployment (BIG-IP and Virtual Services) are defined externally and not tied to specific IPs. Blue/Green consists of creating entirely new deployments and relies on DNS. However, if replacing, upgrading or migrating instances in existing deployments, and you want to re-use the BIG-IP instance’s backup file, it is not possible as a new DHCP-based instance’s IPs address will not match the backup (UCS) file. This new option allows you to assign static IP addresses, which enhances the ability to migrate or replace instances in an existing deployment.

Experimental templates in 4.0.0

In addition to minor enhancements and fixes, this maintenance release contains the following:

• New experimental template for clustered BIG-IQ License Manager deployments
  This new experimental template deploys clustered (hot standby) BIG-IQ license managers in a highly available configuration across Amazon Availability Zones, using BYOL (bring your own license) licensing. The BIG-IQ VE can detect Availability Zone failure and automatically shift public traffic to the BIG-IQ in the Availability Zone that is unaffected.

Released on 11/1/18

v3.3.1

In addition to minor enhancements and fixes, this maintenance release contains the following:

• New README files in each supported GitHub directory
  In order to help users browse to the appropriate template, we added README files to each directory. These files contain information on the template types available in each subsequent directory.

• Updated supported auto scale example diagrams
  Updated the configuration example diagrams for the auto scale README files to accurately depict the scenario in the template (for example, the diagrams now accurately show BIG-IQ or BIG-IP DNS where applicable).

• Corrected an issue in Experimental auto scale (Front-end via NLB) templates
  Fixed a double curly braces syntax error in custom-config.sh in the auto scale via NLB templates.

Released on 9/20/18

v3.3.0

Supported templates in release 3.3.0

In addition to minor enhancements and fixes, this release contains the following:

• Auto Scale LTM and WAF templates using DNS now supported
  The auto scale templates for BIG-IP LTM and WAF (ASM) using DNS are now fully supported by F5. By virtue of a DNS request made by the client to the DNS provider (currently only BIG-IP DNS), traffic flows to the current BIG-IP VE (cluster) members directly, which then process traffic to the application servers, allowing you to deploy without using an AWS Elastic Load Balancer (ELB).

• Support for BIG-IP version 13.1.1
  This release introduces support for BIG-IP version 13.1.1 across templates. In this release, for BYOL and BIG-IQ licensing, the BIG-IP image names have changed (previous options were Good, Better, and Best). Now you choose a BIG-IP VE image based on whether you need LTM only or All modules available (including WAF, ASM, etc.), and if you need 1 or 2 boot locations. Use 2 boot locations if you expect to upgrade the BIG-IP VE in the future. If you do not need room to upgrade (if you intend to create a new instance when a new version of BIG-IP VE is released), use an image with 1 boot location.

Experimental templates in release 3.3.0

In addition to minor enhancements and fixes, this release contains the following:

• Standalone BIG-IQ License Manager template
  This experimental template is our first template that deploys a standalone BIG-IQ device. This template launches a 2-NIC deployment of a BIG-IQ License Manager VE in an Amazon Virtual Private Cloud, using a BYOL (bring your own license) BIG-IQ License Manager License. You chose the type of licensing pool you want on the BIG-IQ in order to license your BIG-IP VE instances.

Released on 8/28/18

v3.2.0

In addition to minor enhancements and fixes, this release contains the following:

• Cluster 3 NIC – Across AZs template
  These new templates for deployments across AWS Availability zones launch and configure two BIG-IP 3-NIC VEs in a clustered, highly available configuration. If one device becomes unavailable, the standby takes over traffic management duties, ensuring you have the highest level of availability for your applications. There are versions of this template for hourly (pay-as-you-go), BYOL (bring your own license), and licensing using a BIG-IQ device.

• F5 AWS repository structure changes
  In an effort to provide consistency across our different cloud offerings, we have modified the directory structure of the F5 AWS CloudFormation template repository. Now each template has its own folder with an associated readme file specific to the template. Any bookmarked links will likely need to be updated due to this change.

• Standalone and cluster templates have a new default AWS instance size
  All F5 standalone and cluster templates now default to using the AWS m4.xlarge instance size. Previously the default was m3.2xlarge. You can still choose the AWS instance size from the drop-down list in the template, only the default in the CFT has changed.

Released on 6/28/18

v3.1.0

In addition to minor enhancements and fixes, this release contains the following:

• Cluster 3 NIC – Same AZ template
  This new template for same Availability Zone deployments launches and configures two BIG-IP 3-NIC VEs in a clustered, highly available configuration in an Amazon Availability Zone. If one device becomes unavailable, the standby takes over traffic management duties, ensuring you have the highest level of availability for your applications.

• All templates with BIG-IQ licensing now support using a ELA/subscription pool
  The templates now support BIG-IQ licensing using an ELA (Enterprise Licensing Agreement)/subscription pool, which enables self-licensing of BIG-IP virtual editions (VEs). This reduces time and effort and enables the quick delivery of application and security services.

• Templates now allow use of custom BIG-IP image
  The templates now support using a custom BIG-IP image if necessary. You simply add the AMI ID in the Custom Image ID field, and the custom BIG-IP image launches. This allows greater flexibility for your cloud deployments.

Released on 5/17/18

v3.0.1

This minor release was only to update the f5.aws_advanced_ha iApp template (https://github.com/F5Networks/f5-aws-cloudformation/tree/master/iApps)

Released on 4/30/18

v3.0.0

Supported templates additions in 3.0.0

In addition to minor enhancements and fixes, this release contains the following:

• Standalone nNIC (multi-NIC) template fully supported
  The standalone nNIC template has moved from experimental to supported. This template allows you to specify additional network interfaces (up to a total of 8) for your BIG-IP VE. This template is extremely useful when the BIG-IP VE is being used as a traffic controller and requires more than 3 network interfaces.

• Auto scale templates now automatically backup BIG-IP configuration
  All AWS auto scale templates now automatically create a UCS backup of the BIG-IP VE configuration and store it in the AWS S3 bucket created by the template. Backups occur once a day at 12am and 7 days of backups are saved. In the event the system needs to restore from a backup UCS file, it uses the file with the latest timestamp.

• Logging iApp template
  F5 has created an iApp for configuring logging for BIG-IP modules to be sent to a specific set of cloud analytics solutions. The iApp creates logging profiles which can be attached to the appropriate objects (virtual servers, APM policy, and so on) which results in logs being sent to the selected cloud analytics solution, AWS in this case. This iApp now loads on the BIG-IP VE as a part of the template.

Experimental template additions in 3.0.0

• Learning stacks
  The experimental F5 CFT templates now include an option for learning stacks. Learning stacks means that in addition to creating a new networking stack, it also creates all of the appropriate configuration for an example web service. This allows you to create a quick, working demo, or a place to get hands on learning, with a click of a button; everything is created from scratch for you. See https://www.youtube.com/watch?v=jBLMVsM0fM4 for a video of a learning stack template in action.

• Auto scale using DNS (no ELB required)
  These new experimental templates for our auto scale solutions now use a DNS request made by the client to the DNS provider (currently only BIG-IP DNS (formerly BIG-IP GTM) is supported) to direct traffic to the current BIG-IP VE (cluster) members directly, which will then process traffic to the application servers. This allows deployments without using an Elastic Load Balancer (ELB), which saves money and simplifies the configuration.

• Auto scale using BIG-IQ and scaling with PAYG instances
  These new experimental auto scale templates launch at least one BIG-IP VE Local Traffic Manager with a BYOL (bring your own license) license, but then use hourly BIG-IP VEs when auto scale events occur. The templates rely on an existing F5 BIG-IQ device with a pool of 1-4 BYOL licenses. But instead of scaling BYOL BIG-IP VE instances as our other auto scale solutions do, as traffic through the BIG-IP VE or the BIG-IP CPU utilization increases, up to 8 hourly BIG-IP VE LTM instances automatically instantiate to seamlessly handle the load. When traffic or CPU utilization decreases, the hourly BIG-IP VEs are automatically removed.

Released 4/12/18