search

deis / charts

(OBSOLETE) Helm Classic v1 Charts for Deis Workflow
https://deis.com/workflow/
MIT License
45 stars 36 forks source link

chore(workflow-v2.2.0): releasing workflow-v2.2.0(-e2e) #310

Closed smothiki closed 8 years ago

smothiki commented 8 years ago

charts for v2.2.0. Tested by installing deis on v1.2.5 k8s cluster There are some errors from deisrel which I manually corrected. Please take a closer look before giving LGTM

bacongobbler commented 8 years ago

from grokking the chart it appears that there are 18 instances in workflow-v2.2.0 that are using the canary images. We should switch them to git tags. I'm aware that we know this internally, just posting publicly :)

smothiki commented 8 years ago

@bacongobbler updated to git tags

smothiki commented 8 years ago

Fixed

bacongobbler commented 8 years ago

Awesome. Back to testing!

bacongobbler commented 8 years ago

Both @mboersma and I are hanging for a very long time when we attempt to uninstall workflow-v2.2.0 from our cluster:

Uninstall the listed objects? (y/N) y
---> Running `kubectl delete` ...
---> service "deis-builder" deleted

---> service "deis-controller" deleted

---> service "deis-database" deleted

[WARN] Not uninstalling Service deis-logger-redis because of "helm-keep" annotation.
---> Use kubectl to uninstall keeper manifests.

---> service "deis-logger" deleted

---> service "deis-minio" deleted

---> service "deis-monitor-grafana" deleted

---> service "deis-monitor-influxapi" deleted

---> service "deis-monitor-influxui" deleted

---> service "deis-nsqd" deleted

[WARN] Not uninstalling Service deis-registry because of "helm-keep" annotation.
---> Use kubectl to uninstall keeper manifests.

[WARN] Not uninstalling Service deis-router because of "helm-keep" annotation.
---> Use kubectl to uninstall keeper manifests.

---> service "deis-workflow-manager" deleted

---> replicationcontroller "deis-builder" deleted

---> replicationcontroller "deis-controller" deleted

---> replicationcontroller "deis-database" deleted

---> replicationcontroller "deis-logger" deleted

---> replicationcontroller "deis-logger-redis" deleted

---> replicationcontroller "deis-minio" deleted

---> replicationcontroller "deis-monitor-grafana" deleted

---> replicationcontroller "deis-monitor-influxdb" deleted

---> replicationcontroller "deis-nsqd" deleted

---> replicationcontroller "deis-registry" deleted

---> replicationcontroller "deis-router" deleted

---> replicationcontroller "deis-workflow-manager" deleted

---> daemonset "deis-logger-fluentd" deleted

---> daemonset "deis-monitor-telegraf" deleted

The problem seems to be coming from the reigstry-proxy as everything zips by as soon as it's removed.

---> daemonset "registry-proxy" deleted

---> secret "builder-ssh-private-keys" deleted

---> secret "builder-key-auth" deleted

---> secret "django-secret-key" deleted

---> secret "database-creds" deleted

[WARN] Not uninstalling Secret logger-redis-creds because of "helm-keep" annotation.
---> Use kubectl to uninstall keeper manifests.

---> secret "objectstorage-keyfile" deleted

---> secret "deis-router-dhparam" deleted

---> serviceaccount "deis-builder" deleted

---> serviceaccount "deis-controller" deleted

---> serviceaccount "deis-database" deleted

---> serviceaccount "deis-logger-fluentd" deleted

---> serviceaccount "deis-logger" deleted

---> serviceaccount "deis-minio" deleted

---> serviceaccount "deis-monitor-telegraf" deleted

---> serviceaccount "deis-nsqd" deleted

---> serviceaccount "deis-registry" deleted

---> serviceaccount "deis-router" deleted

---> serviceaccount "deis-workflow-manager" deleted

[WARN] Not uninstalling Namespace deis because of "helm-keep" annotation.
---> Use kubectl to uninstall keeper manifests.

---> Done

Total time is ~2 minutes in comparison to the old ~15 seconds.