This repository has been archived by the owner on Jan 12, 2024. It is now read-only.
Releases: Nike-Inc/cerberus-lifecycle-cli
Releases · Nike-Inc/cerberus-lifecycle-cli
v3.0.1
v3.0.0
v2.3.0
v2.2.0
v2.1.0
- Consul configuration: setting
reconnect_timeout
andreconnect_timeout_wan
to 16 hours. Any value lower than the default of 72 hours seems better when we're using CloudFormation for deploys. 16 hours is twice the minimum of 8 hours. - Sorting CLI
--help
output by command name to make it easier to read.
v2.0.0
v1.7.0
v1.6.0
v1.5.0
Vault updates no longer incurs up to (10 Seconds * instance count in the ASG) of downtime during deploys.
- Configured the Vault health check to report standby nodes as healthy so all instances get registered as healthy to the ELB.
- Configured the Vault Security Group to allow request forwarding that was introduced since the initial release of Cerberus to work properly with new ELB configuration
To enable this run run the following commands
cerberus -e xxxx -r yyyy update-stack --stack-name base --overwrite-template --owner-group foo
cerberus -e xxxx -r yyyy update-stack --stack-name vault --overwrite-template --owner-group foo
A bonus side effect of these changes, under high load the ELB doesn't freak out and drop the leader returning 503s.
v1.4.1
- Consul config change: adding 'acl_enforce_version_8: false' needed before we upgrade to 0.8.x (or we need to update our policies https://www.consul.io/docs/guides/acl.html#complete-acl-coverage-in-consul-0-8). Run
update-consul-config
to apply this config change before the new Consul AMI is deployed Nike-Inc/cerberus-consul-puppet-module#10 - Consul client also needs the same config change