profile
viewpoint

bradbeam/ads 0

CoreDNS plugin to block ads. Inspried by PiHole.

bradbeam/alertmanager 0

Prometheus Alertmanager

bradbeam/auditd 0

Install and configure user mode auditd tools

bradbeam/bird 0

Calico's fork of the BIRD protocol stack

bradbeam/bootkube 0

bootkube - Launch a self-hosted Kubernetes cluster

bradbeam/calico 0

Cloud native application connectivity and network policy

PullRequestReviewEvent
PullRequestReviewEvent

push eventbradbeam/redskyops-controller

Jeremy Gustie

commit sha 4b9ac82410af09b97679468bc47109330633569f

Fix authorize cluster Helm output to match the current version of the chart

view details

Jeremy Gustie

commit sha 9711d5363a6fa4e447dce0419095f88f6af8c5a4

Bump the Helm chart using a workflow dispatch

view details

Jeremy Gustie

commit sha 7937ee63802849c434eeaf089d8122646c56e46d

Merge pull request #437 from thestormforge/helm-changes Helm changes

view details

Jeremy Gustie

commit sha 47586beb814ba56e48d2b1c51ef12670ea625dfc

Use GitHub CLI for Helm chart release

view details

Jeremy Gustie

commit sha bf6005a9a4dafbf00f8e5b7f285cff99c5291361

Merge pull request #439 from thestormforge/use-gh Use GitHub CLI for Helm chart release

view details

Brad Beam

commit sha c85f8beb6bbe93f5c78d7986184a89df388545ad

refactor: Use MODE environment variable instead of args This changes the entrypoint to use the MODE variable to trigger the action that should be performed. This allows us to clean up the creation of the startup task and simplify some of the logic. Signed-off-by: Brad Beam <brad.beam@stormforge.io>

view details

push time in 7 days

Pull request review commentthestormforge/optimize-controller

refactor: Use MODE environment variable instead of args

 #!/bin/sh set -e -case "$1" in-  prometheus)-    shift+waitFn() { :; } -    cat <<-EOF >helm.yaml-		apiVersion: konjure.carbonrelay.com/v1beta1-		kind: HelmGenerator-		metadata:-		  name: prometheus-		releaseName: optimize-${NAMESPACE}-prometheus-		releaseNamespace: ${NAMESPACE}-		chart: ../prometheus-		EOF--    export HELM_CONFIG=$(cat helm.yaml | base64 -w0)--    waitFn() {-      # Wait on {{ releaseName }}-server-      kubectl wait --for condition=Available=true --namespace "${NAMESPACE}" --timeout 5m deployment.apps "optimize-${NAMESPACE}-prometheus-server"-    }-  ;;-  *)-    waitFn() { :; }-  ;;-esac+while [ "$#" != "0" ] ; do+  case "$1" in+    prometheus)+      shift++      cat <<-EOF >helm.yaml+			apiVersion: konjure.carbonrelay.com/v1beta1+			kind: HelmGenerator+			metadata:+			  name: prometheus+			releaseName: optimize-${NAMESPACE}-prometheus+			releaseNamespace: ${NAMESPACE}+			chart: ../prometheus+			EOF++      export HELM_CONFIG=$(cat helm.yaml | base64 -w0)

Yeah, that's the reason for installing coreutils. I had a similar confused look because I hadn't noticed it either in testing. I put some details in the commit message for https://github.com/thestormforge/optimize-controller/pull/438/commits/80da0f8fe0616d353212851f6de7e2a26e0dd8ed

bradbeam

comment created time in 7 days

PullRequestReviewEvent
PullRequestReviewEvent

push eventbradbeam/redskyops-controller

Brad Beam

commit sha a6f7404bf54d968f83c60d207aee0d03e9bcdd86

refactor: Use MODE environment variable instead of args This changes the entrypoint to use the MODE variable to trigger the action that should be performed. This allows us to clean up the creation of the startup task and simplify some of the logic. Signed-off-by: Brad Beam <brad.beam@stormforge.io>

view details

push time in 8 days

PR opened thestormforge/optimize-controller

refactor: Use MODE environment variable instead of args

This is mostly whitespace changes; the handling of script args moves to the beginning of the script. The initial loop to handle args is provided for backwards compatibility with older setup tasks that make use of prometheus $(MODE).

+70 -62

0 comment

2 changed files

pr created time in 8 days

push eventbradbeam/redskyops-controller

Brad Beam

commit sha ebf2ba4109d7b938d6d011c71a0c0dd1d5d4eb18

refactor: Use MODE environment variable instead of args This changes the entrypoint to use the MODE variable to trigger the action that should be performed. This allows us to clean up the creation of the startup task and simplify some of the logic. Signed-off-by: Brad Beam <brad.beam@stormforge.io>

view details

push time in 8 days

push eventbradbeam/redskyops-controller

Brad Beam

commit sha 4271b3d46ea519263beb48d1e84621c1b5ea9bc6

refactor: Use MODE environment variable instead of args This changes the entrypoint to use the MODE variable to trigger the action that should be performed. This allows us to clean up the creation of the startup task and simplify some of the logic. Signed-off-by: Brad Beam <brad.beam@stormforge.io>

view details

push time in 8 days

create barnchbradbeam/redskyops-controller

branch : entrypointmode

created branch time in 8 days

PullRequestReviewEvent

pull request commentthestormforge/optimize-controller

feat: Explicitly use prometheus chart in experiment generation

This needs some work ( currently setup task entry point requires $(MODE) as an argument )

bradbeam

comment created time in 11 days

delete branch bradbeam/redskyops-controller

delete branch : entrypoint

delete time in 12 days

push eventthestormforge/optimize-controller

Brad Beam

commit sha 10bd2969d3bd30621d90ca3de64e5024ed45eaa1

fix: Wait for deployment resources in prometheus setup task This fixes an issue where we would create a deployment but then wait for pods to be ready. This was flakey because we would sometimes attempt to check/wait for the pods to be ready before kube had a chance to actually to create the pod. Signed-off-by: Brad Beam <brad.beam@stormforge.io>

view details

Brad Beam

commit sha 86fd9ea6cb4df70b625945621ed03095e182f3b2

Merge pull request #436 from bradbeam/entrypoint fix: Wait for deployment resources in prometheus setup task

view details

push time in 12 days

PR merged thestormforge/optimize-controller

fix: Wait for deployment resources in prometheus setup task

This fixes an issue where we would create a deployment but then wait for pods to be ready. This was flakey because we would sometimes attempt to check/wait for the pods to be ready before kube had a chance to actually to create the pod.

Signed-off-by: Brad Beam brad.beam@stormforge.io

+10 -12

0 comment

1 changed file

bradbeam

pr closed time in 12 days

PR opened thestormforge/optimize-controller

fix: Wait for deployment resources in prometheus setup task

This fixes an issue where we would create a deployment but then wait for pods to be ready. This was flakey because we would sometimes attempt to check/wait for the pods to be ready before kube had a chance to actually to create the pod.

Signed-off-by: Brad Beam brad.beam@stormforge.io

+10 -12

0 comment

1 changed file

pr created time in 12 days

create barnchbradbeam/redskyops-controller

branch : entrypoint

created branch time in 12 days

PR opened thestormforge/optimize-controller

feat: Allow ignoring of unschedulable condition via annotation

This adds in a new annoation, stormforge.io/skip-unschedulable to allow us to ignore unschedulable status.

Signed-off-by: Brad Beam brad.beam@stormforge.io

+5 -1

0 comment

1 changed file

pr created time in 20 days

create barnchbradbeam/redskyops-controller

branch : ignoreunsched

created branch time in 20 days

PullRequestReviewEvent
PullRequestReviewEvent
PullRequestReviewEvent
PullRequestReviewEvent

delete branch bradbeam/redskyops-controller

delete branch : readychecktimeout

delete time in 20 days

push eventthestormforge/optimize-controller

Brad Beam

commit sha 30532f125c2b0c1e0290dd6250f2766652ce259c

feat: Increase timeout on wait for setup task to complete Signed-off-by: Brad Beam <brad.beam@stormforge.io>

view details

Brad Beam

commit sha c3bb4a33bc5ac7a532ad73362f52237cccb264d0

Merge pull request #429 from bradbeam/readychecktimeout feat: Increase timeout on wait for setup task to complete

view details

push time in 20 days

create barnchbradbeam/redskyops-controller

branch : expgenpromchart

created branch time in 20 days

more