0.104.0 [pritunl] Add feature to expose multiple TCP and UDP ports to pritunl

what

1. [pritunl] exposing multiple ports via variable that accepts comma delimited list of ports
2. [pritunl] fix variable without default, otherwise it was failing with

could not deduce `environment:` block, configuring only .Environment.Name. error: failed to read pritunl.yaml.part.0: reading document at index 1: yaml: line 95: did not find expected key
failed to read pritunl.yaml: reading document at index 1: yaml: line 95: did not find expected key

why

1. so pritunl can serve multiple vpn configurations

Those changes are tested when PRITUNL_VPN_PORT was not configured, when it had the same value as default and for two ports.
Only change is that we need to enumerate services added.

0.103.0 [sentry] Sentry has new SENTRY_IMAGE_REPOSITORY variable

what

1. [sentry] New variable SENTRY_IMAGE_REPOSITORY allowing to change default image repository

why

1. Sentry 9 is not going to be develop further but we still want to support it. Thus using different image repository allows us to fix sentry issues.

0.102.1 [nginx-ingress] Remove invalid url link from alert #245

what

1. [nginx-ingress] removed invalid link from 5xx alert

why

1. invalid link is of no use

0.102.0 [nginx-ingress] Added alerts for nginx-ingress error rate on 4xx and 5xx status codes

what

1. [nginx-ingress] Added alert for 5xx and 4xx error rate

why

1. To detect problems in traffic routed via nginx-ingress

0.101.0 [sentry] Alerting improved

what

1. [sentry] added alarm on time to process, error rate and 5xx errors

why

1. Improving sentry observability

0.100.0 [fluentd-kubernetes-aws] Change metrics names and improve alerts by introducing absent checks

What

• Reflect new metric names in alerts
• Added checks for absent metrics
• Change job to match servicemonitor configured

Why

• Alerts were monitoring nonexistent metrics so they would never be triggered

0.99.1 quotes fix

what

1. [fluentd-elasticsearch-aws] quotes fix

why

1. helmfile was broken

0.99.0 [fluentd] throttle environment variables added

what

1. [fluentd-elasticsearch-aws] throttling values passed using environment variables

why

1. to allow tune throttling easier

0.98.0 [sentry] affinity added to spread web and workers across nodes

what

1. [sentry] affinity added to spread web and workers across nodes

why

1. to reduce downtimes in case of node down

0.97.0 [kiam] version upgrade to version that supports IMDS v2

what

1. [kiam] Kiam 3.6-rc1 accepts by default requests to IMDSv2

why

1. New versions of AWS SDK can use IMDSv2 which is blocked prior to kiam 3.6-rc0 version. This creates a lot of alarms based on kiam_metadata_proxy_requests_blocked_total which are false positives as clients can fallback to IMDSv1.

Ref:

• Instance Metadata Service Version 2 (IMDSv2) https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/configuring-instance-metadata-service.html
• uswitch/kiam#359
• uswitch/kiam#381