civo / kubernetes-marketplace Goto Github PK

Issue with the pods in cattle-system namespaces when the rancher UI is accessed.

I was looking for an application server like Tomcat and did not find it. Even if I wanted to propose adding one (Tomcat or another that I was more prepared to be maintainer for), this raised a few questions:

• is there any reason app servers would not be accepted? (I don't find mention of any in issues, so clearly the demand is low)
• what category would you have such a container listed as? Or might you create a new one, like "application servers"?
• are only containers for open source apps allowed? What if a commercial vendor offers a free dev edition of a container?

Thanks

This issue is a:
[ ] Bug Report
[ ] Enhancement suggestion
[*] Feature request

Enhancement suggestions / feature requests

If you are requesting an enhancement to a feature already present, or a new feature, please outline below the feature you would like to see, and how it would benefit you and the wider Civo Kubernetes userbase:

Example chart
https://bitnami.com/stack/kafka/helm

This issue is a:
[ ] Bug Report
[x] Enhancement suggestion
[ ] Feature request

In deploying a new cluster, I noticed that there's Ghost, Wordpress and the like in the "Management" category of the Marketplace.

Would it be better to have a more defined category of "management" (as in managing your cluster) and add a category/categories for CMSes and the like such as "content"? I thought I'd open up an issue here to allow for discussion between all potential users. If the number of categories total is an issue, I'd suggest moving some of the more cluster management-oriented apps (Rancher, Kubenav, Dashboard) to "monitoring" and renaming the management category.

For discussion, my suggested category listing, assuming adding new categories is ok, is:

• Storage
• Database
• Management
• Monitoring
• CI/CD
• Architecture
• Security
• Content (new)

This issue is a: Bug Report

Due to security reasons of your company, i am adding a link that will ask for request access to view this report.

Bug Report

Minio Pod is failing to create with the following error (pod coming up on the master node which has no NodeD):

  Type     Reason              Age                From                       Message
  ----     ------              ----               ----                       -------
  Normal   Scheduled           2m50s              default-scheduler          Successfully assigned default/minio-65c68b8ffb-wj6lb to kube-master-449c
  Warning  FailedAttachVolume  67s (x3 over 69s)  attachdetach-controller    AttachVolume.Attach failed for volume "minio-pv" : node "kube-master-449c" has no NodeID annotation
  Warning  FailedMount         47s                kubelet, kube-master-449c  Unable to mount volumes for pod "minio-65c68b8ffb-wj6lb_default(dacf858f-5418-4f8f-af81-d46a4489d78c)": timeout expired waiting for volumes to attach or mount for pod "default"/"minio-65c68b8ffb-wj6lb". list of unmounted volumes=[minio-data]. list of unattached volumes=[minio-data default-token-c729p]
  Warning  FailedAttachVolume  4s (x5 over 65s)   attachdetach-controller    AttachVolume.Attach failed for volume "minio-pv" : rpc error: code = NotFound desc = ControllerPublishVolume: the volume minio-data not exists

This issue is a:
[ ] Bug Report
[ ] Enhancement suggestion
[*] Feature request

If the issue concerns an existing Marketplace application, please tag the maintainer of the application (found in manifest.yaml) below:

Enhancement suggestions / feature requests

If you are requesting an enhancement to a feature already present, or a new feature, please outline below the feature you would like to see, and how it would benefit you and the wider Civo Kubernetes userbase:

Bug Reports

If the issue is a Bug Report, please detail the following:

• Expected behaviour
• Encountered behaviour
• The steps to recreate the behaviour that you are reporting

If you have logs or screenshots, please include them below.

This issue is a:
[ ] Bug Report
[ ] Enhancement suggestion
[ ] Feature request

If the issue concerns an existing Marketplace application, please tag the maintainer of the application (found in manifest.yaml) below:

Enhancement suggestions / feature requests

If you are requesting an enhancement to a feature already present, or a new feature, please outline below the feature you would like to see, and how it would benefit you and the wider Civo Kubernetes userbase:

Bug Reports

If the issue is a Bug Report, please detail the following:

• Expected behaviour
• Encountered behaviour
• The steps to recreate the behaviour that you are reporting

If you have logs or screenshots, please include them below.

This issue is a:
[ ] Bug Report
[ ] Enhancement suggestion
[* ] Feature request

If the issue concerns an existing Marketplace application, please tag the maintainer of the application (found in manifest.yaml) below:

Enhancement suggestions / feature requests

If you are requesting an enhancement to a feature already present, or a new feature, please outline below the feature you would like to see, and how it would benefit you and the wider Civo Kubernetes userbase:

Bug Reports

If the issue is a Bug Report, please detail the following:

• Expected behaviour
• Encountered behaviour
• The steps to recreate the behaviour that you are reporting

If you have logs or screenshots, please include them below.

This issue is a:
[ ] Bug Report
[x] Enhancement suggestion
[ ] Feature request

Enhancement suggestions / feature requests

A new category "CMS" would be necessary to allow applications like Wordpress (see PR #91)

While trying to use longhorn for persistent storage in my civo k8s cluster, I encountered an issue where the pod was failing to mount the volume:

3m43s Warning FailedAttachVolume pod/plaindb-86dcdfb76-ppcl9 AttachVolume.Attach failed for volume "plaindb-volume" : attachdetachment timeout for volume plaindb-volume
2m8s Warning FailedMount pod/plaindb-86dcdfb76-ppcl9 Unable to attach or mount volumes: unmounted volumes=[plaindb-volume], unattached volumes=[default-token-vwkvg plaindb-volume]: timed out waiting for the condition

After a lot of troubleshooting it turned out that PV creation was causing the issue. So, when using longhorn with civo, we just need to create a claim and use that claim in the pod / deployment or whatever manifest you want to use. Creating PV beforehand causes some issues with the volume creation in the backend.

Can you please remove pv.yaml example from longhorn/post_install.md file, or at least put a note that PV creation is not needed and might cause issues.

This issue is a:

• Bug Report
• Enhancement suggestion
• Feature request

The current cert-manager installation is outdated, and currently doesn't work as intended in the current K3S version. A companion PR, GH-160 is created to use the official one-line YAML installation instead of the usual Jetstack Helm chart.

This issue is a:
[ ] Bug Report
[X] Enhancement suggestion
[ ] Feature request

When currently looking at the marketplace listing there is no indication of the minimum cluster configuration required or preferred. Take prometheus-operator that is quite demanding in terms of resources. If you just have one node in configuration small then not much more will be able to run.

As a first step we could add an estimation to the description. As a second step we could add some entries to manifest.yml where minimum requirements are stored. Based on these the Civo dashboard could show an estimate of the resource consumption based on installed apps.

In general this will be most helpful to new users. As I think that Civo service also targets newcomers and others that do not breathe k8s concepts such hints might be useful.

This issue is a:
[ ] Bug Report
[x] Enhancement suggestion
[ ] Feature request

About RabbitMQ you can use the rabbitmq operator:
https://www.rabbitmq.com/kubernetes/operator/operator-overview.html

Currently, civo installation exposes the standard HTTP and amqp port. I would suggest using the HTTPS and AMQPS

This issue is a:

• Bug Report
• Enhancement suggestion
• Feature request

If the issue concerns an existing Marketplace application, please tag the maintainer of the application (found in manifest.yaml) below:

Enhancement suggestions / feature requests

If you are requesting an enhancement to a feature already present, or a new feature, please outline below the feature you would like to see, and how it would benefit you and the wider Civo Kubernetes userbase:

the repo of the helm chart needs updating as it has moved repos now
helm/charts#23505 (comment)

Bug Reports

If the issue is a Bug Report, please detail the following:

• Expected behaviour
• Encountered behaviour
• The steps to recreate the behaviour that you are reporting

If you have logs or screenshots, please include them below.

• [ ]

This issue is a:

• Bug Report
• Enhancement suggestion
• Feature request

We have some offending Applications that still uses deprecated API namespaces, from my discovery, the following WILL NOT work on K3s 1.20+

• litmuschaos (uses apiextensions.k8s.io/v1beta1, deprecated in favor of apiextensions.k8s.io/v1)
• bitwardenrs (uses extensions/v1beta1 Ingress classes. Will not work on Kubernetes 1.22)
• rancher (example uses deprecated extensions/v1beta1 Ingress class for post install guide)

This issue is a: Feature request

A logging solution like EFK or Graylog will be very help full.

This issue is a:
[x] Feature request

Enhancement suggestions / feature requests

It would be good to have the official nfs-provisioner (https://github.com/kubernetes-incubator/external-storage/tree/master/nfs)

This issue is a:
[ ] Bug Report
[ ] Enhancement suggestion
[*] Feature request

If the issue concerns an existing Marketplace application, please tag the maintainer of the application (found in manifest.yaml) below:

Enhancement suggestions / feature requests

If you are requesting an enhancement to a feature already present, or a new feature, please outline below the feature you would like to see, and how it would benefit you and the wider Civo Kubernetes userbase:

This helm chart:
https://github.com/hashicorp/vault-helm
https://github.com/hashicorp/vault-helm/blob/master/values.yaml

Example of other apps using this chart:
https://docs.gitlab.com/ee/user/clusters/applications.html#install-vault-using-gitlab-cicd

This issue is a:

• Bug Report
• Enhancement suggestion
• Feature request

Bug Reports

• Expected behaviour
  When running $ civo kubernetes applications ls CLI command, I'm expecting a list of all Civo Marketplace apps with Plans column contains values like 5GB, 10GB, 15GB.

• Encountered behaviour
  What I noticed is, Linkerd plans are strings with the following values: Linkerd Minimal, Linkerd & Jaeger, Linkerd with Dashboard, Linkerd with Dashboard & Jaeger.

  Is this a mistake?

  Because of this, the Plans column became too wide and causing double rows problem to other apps when the viewport is small e.g. a normal laptop screen (not an external display).

  If this bug report is incorrect/false alarm, feel free to close this.

• The steps to recreate the behaviour that you are reporting
  Run $ civo kubernetes applications ls CLI command

  

This issue is a:

• Bug Report
• Enhancement suggestion
• [ x] Feature request

If the issue concerns an existing Marketplace application, please tag the maintainer of the application (found in manifest.yaml) below:

Enhancement suggestions / feature requests

Many Java applicattions that built requires a in cluster repository manager for quicker build, Snoatype Nexus is one such repository manager that helps in quicker and faster builds.

It will be great to have one installed via market palce for the applications to use them. E.g. Tekton Pipelines that builds Java can use it via `MAVEN_MIRROR_URL1

This issue is a:
[ ] Bug Report
[x] Enhancement suggestion
[ ] Feature request

If the issue concerns an existing Marketplace application, please tag the maintainer of the application (found in manifest.yaml) below: @alexellis

Enhancement suggestions / feature requests

Newer cert-manager version e.g. 0.14. 0.11 is getting quite old now.

This issue is a:
[ ] Bug Report
hlw, i am beta tester,
whenever i am creating an instance or installing apps, this will take me to the Error message look like "Sorry for trouble if persist contact on mail"
but this behaviour is only in the frontend but the instance is creating in the backgroung.

Bug Reports

If the issue is a Bug Report, please detail the following:

• Expected behaviour
  Try to create an instance or install apps, some times you will definitely get this error.

This issue is a:
[ ] Bug Report
[ ] Enhancement suggestion
[ * ] Feature request

If the issue concerns an existing Marketplace application, please tag the maintainer of the application (found in manifest.yaml) below:

Enhancement suggestions / feature requests

Adding Spinnaker to the marketplace might be a nice touch

This issue is a:
[ ] Bug Report
[ ] Enhancement suggestion
[ * ] Feature request

If the issue concerns an existing Marketplace application, please tag the maintainer of the application (found in manifest.yaml) below:

Enhancement suggestions / feature requests

While exploring civo kubernetes service. What I really missed is deploying any security related tools on k8s through marketplace.
As a start, I would like to add tools like kube-scan and polaris to the marketplace.

This issue is a:

• Bug Report
• Enhancement suggestion
• Feature request

Currently we only have Longhorn, however, if you're using Rancher, you'll be able to deploy anything other than Longhorn (such as OpenEBS, which is a much more better solution since it supports RWX (Read-Write-Many) vs Longhorn).

This issue is a:
[x] Bug Report
[ ] Enhancement suggestion
[ ] Feature request

If the issue concerns an existing Marketplace application, please tag the maintainer of the application (found in manifest.yaml) below:
@javicv

Bug Reports

Wordpress installation is not triggered correctly. When you select "Wordpress" from marketplace applications, it's shown as installed with its dependencies but if check the cluster neither of them are installed (longhorn, mariadb, wordpress)

In a new cluster I tried to install MariaDB and it triggers Longhorn's installation and MariaDB is installed afterwards, then I select Wordpress from the marketplace and the application isn't installed, I can't see the pvc, deployment, service or ingress in my cluster with kubectl.

Is there something wrong with the application template or is it caused by the marketplace installer?

This issue is a:
[ ] Bug Report
[ ] Enhancement suggestion
[*] Feature request

If the issue concerns an existing Marketplace application, please tag the maintainer of the application (found in manifest.yaml) below:

Enhancement suggestions / feature requests

If you are requesting an enhancement to a feature already present, or a new feature, please outline below the feature you would like to see, and how it would benefit you and the wider Civo Kubernetes userbase:

This issue is a:
[ ] Bug Report
[ ] Enhancement suggestion
[X ] Feature request

Enhancement suggestions / feature requests

If you are requesting an enhancement to a feature already present, or a new feature, please outline below the feature you would like to see, and how it would benefit you and the wider Civo Kubernetes userbase:

I would like to add the app https://docs.dapr.io/ to Civo marketplace, its a proposal, which I would like to take up and maintain for Distributed Application Runtime. Dapr is an open source and vendor neutral project, its also an event-driven, portable runtime for building microservices on cloud.

I am interested to know the complexity in adding an app and its experience in Civo marketplace.
Also I am new to Dapr, this will give us some experience how to to build resilient, microservice stateless and stateful applications that run on Civo.

It'd be great if there were issue templates to help identify a problem

And PR templates with a check-list that included things like the contributing guide being agreed to:

• I agree to maintain this submission
• To the best of my knowledge this is an original work and doesn't contain copyrighted materials
• I have tested the application (screenshot or text attached below)

Please give the maintainer's GitHub profile below (if not yourself):

• Username: @alexellis

This issue is a:

• Bug Report
• Enhancement suggestion
• Feature request

Enhancement suggestions / feature requests

Flux2 is a widely used, extensible open-source continuous delivery solution powered by GitOps Toolkit.

This issue is a:
[ ] Bug Report
[ ] Enhancement suggestion
[x] Feature request

If the issue concerns an existing Marketplace application, please tag the maintainer of the application (found in manifest.yaml) below:

Enhancement suggestions / feature requests

If you are requesting an enhancement to a feature already present, or a new feature, please outline below the feature you would like to see, and how it would benefit you and the wider Civo Kubernetes userbase:

Well, CockroachDB is a great evolving database that has great hybrid and multi-cloud compatibility and also offers a service called Managed CockroachDB for Kubernetes that allows the users to run the database worry-free and can request for operations using a convenient UI. CockroachDB is also a common face in the Hackathons. So, it has got a great amount of exposure to young techies as well.

Bug Reports

If the issue is a Bug Report, please detail the following:

• Expected behaviour
• Encountered behaviour
• The steps to recreate the behaviour that you are reporting

If you have logs or screenshots, please include them below.

@JasonMorgan So right now there are three apps in the production can we have Plans for them and have only single app ? then in the install.sh based on the value of plan we run the respective helm install.

So we can have something like this in the PLANS section

plans:
  - label: "Linkerd"
    configuration:
      LINKERD
        value: linkerd
  - label: "linkerd+jaeger"
    configuration:
      LINKERD
        value: linkerdjaeger
  - label: "Linkerd+viz"
    configuration:
      LINKERD:
        value: linkerdviz

now in the install.sh we can have below

#!/bin/bash

curl -sL https://run.linkerd.io/install | sh

export PATH=$PATH:$HOME/.linkerd2/bin
linkerd check --pre && linkerd install | kubectl apply -f -
if [$LINKERD == "linkerd"]
then
   linkerd check || exit 1
elif [ $LINKERD == "linkerdjaeger" ] 
then 
    linkerd check && linkerd jaeger install | kubectl apply -f -
    linkerd check || exit 1
else 
    linkerd check && linkerd viz install | kubectl apply -f -    
    linkerd check || exit 1
fi 

In this way, we can have options for installations within the same application

@andyjeffries thoughts ?

This issue is a:

• Bug Report
• Enhancement suggestion
• Feature request

Enhancement suggestions / feature requests

As microservices and Istio application developer, I wish to have Gloo Edge to be installed and configured as part of my civo cluster.

As Gloo Edge exposes API Gateway for Kubernetes based microservices, it need to bind to 80 and 443 ports. This requirement makes it mutually exclusive with Traeifk which is by default installed in k3s clusters. As part of this application install, it will first uninstall Traefik before installing Gloo Edge

This issue is a:
[ ] Bug Report
[x] Enhancement suggestion
[ ] Feature request

If the issue concerns an existing Marketplace application, please tag the maintainer of the application (found in manifest.yaml) below: N/A

Enhancement suggestions / feature requests

Please supply a newer Helm version. If not 3.1 at least 2.16 which is more commonly installed than 2.14 at this point in time.

@alejandrojnm is trying to implement Hasura as a marketplace application, but it requires a PostgreSQL database.

First thoughts

My original thoughts were to allow passing of configuration items from one application to another. e.g.

configuration:
  DB_USERNAME:
    label: "Access key"
    value: "CIVO:APPLICATION:POSTGRES:ROOT_USER_NAME"

However, I don't think it's nice if Application B can pull the authentication details from Application A.

New plan

So now I'm thinking about a slightly nicer interface where applications can provide resources. So PostgreSQL would define something like this in the manifest.yaml:

resources:
  provide:
    - database

and it would have a new resource.sh script which takes two command line positional parameters, the action of either create or delete (for later uninstalling) and a name, which will by default by the application followed by the resource name lowercased. It will output a single line containing the details required to access the resource (usually a URL):

$ resource.sh create hasura-postgres_db
postgres://RANDOM_USERNAME:[email protected]:5432/hasura-postgres_db

Applications that require a database would request one by doing:

dependencies:
  - postgres
resources:
  require:
    POSTGRES_DB:
      application: postgres
      resource: database

The output of the resource.sh script would then be injected in to $POSTGRES_DB the same as if it had been defined as a configuration setting normally.

Would welcome any feedback from anyone writing or thinking about writing marketplace applications.

When longhorn is installed in a fresh cluster, it is not marked as the default storageprovisioner. It appears to be annoated correctly in the marketplace app yaml:

But when a new cluster is created that includes the longhorn app:

civo kubernetes create --size=g2.medium --nodes=3 --applications=cert-manager,Longhorn,metrics-server --wait --save mynewcluster

The storageclass.kubernetes.io/is-default-class annotation is not set. Manually patching it after installation does seem to work.

kubectl patch storageclass longhorn -p '{"metadata": {"annotations":{"storageclass.kubernetes.io/is-default-class":"true"}}}'

This issue is a:
[ ] Bug Report
[ ] Enhancement suggestion
[*] Feature request

If the issue concerns an existing Marketplace application, please tag the maintainer of the application (found in manifest.yaml) below:

Enhancement suggestions / feature requests

If you are requesting an enhancement to a feature already present, or a new feature, please outline below the feature you would like to see, and how it would benefit you and the wider Civo Kubernetes userbase:

This issue is a:

• Bug Report
• Enhancement suggestion
• Feature request

If the issue concerns an existing Marketplace application, please tag the maintainer of the application (found in manifest.yaml) below:

@realshuting @JimBugwadia

Enhancement suggestions / feature requests

Kyverno installs the "Default Pod Security Standard" Policies with the validation failure action "audit" as default. I think it's a common use case that users want to install also the "Restricted Pod Security Standard" Policies. It's also common that users want to enforce this policies instead of auditing.

I noticed that it is possible to define parameters in form of selection in the marketplace application. So I would like to create a PR to add a selection for the "Pod Security Standard" and the "Validation Failure Action".

This issue is a:
[ ] Feature request

Enhancement suggestions / feature requests

Add TeamCity (and maybe other CI/CD tools) to the CI/CD section.

https://www.jetbrains.com/help/teamcity/setting-up-teamcity-for-kubernetes.html#Kubernetes+Cloud+Profile+Configuration

This issue is a:
[ ] Bug Report
[ ] Enhancement suggestion
[ * ] Feature request

Enhancement suggestions / feature requests

Can we add Factorio so I can spin up a server for my friend group on Civo and write a blog post about it?

I think it would be a nice community addition. You could even have a community-run Factorio server on Civo infrastructure if you fancy it :)

https://hub.helm.sh/charts/stable/factorio

None of the categories seems to make sense for actual end-user applications, such as Wordpress, Mediawiki, or Nextcloud. "Application" is pretty board, but trying to make a full taxonomy for all possibilities also seems unnecessary.

This issue is a:
[ ] Bug Report
[x] Enhancement suggestion
[ ] Feature request

Enhancement suggestions / feature requests

If you are requesting an enhancement to a feature already present, or a new feature, please outline below the feature you would like to see, and how it would benefit you and the wider Civo Kubernetes userbase:

I have been wondering about the possibility of getting the K8S dashboard installed and running as an extension on the marketplace. It's quite straight forward to install, the only issues is the proxy that is required to access it. I can create the marketplace addon to install the dashboard, but how feasible is it to get the proxy listening on the public IP address or elsewhere?

This link is a good resource regarding the dashboad: https://www.replex.io/blog/how-to-install-access-and-add-heapster-metrics-to-the-kubernetes-dashboard

Update Portainer

This issue is a:

• Bug Report
• Enhancement suggestion
• Feature request

Rancher seems to error out with HTTP_TOO_MANY_REDIRECTS despite a fresh deployment or not when using the NGINX Ingress controller.

Reproduction steps:

• Install NGINX Ingress from Kubernetes' Helm chart
• Install Rancher from the Civo Marketplace

Expected: Routing is working.
Current Behavior: Browser returns HTTP_TOO_MANY_REDIRECTS

This issue is a:
[X ] Bug Report
[ ] Enhancement suggestion
[ ] Feature request

@akosveres

The manifest of prometheus operator app specifies helm as a dependency that is not needed.

This issue is a:

• Bug Report
• Enhancement suggestion
• Feature request

Enhancement suggestions / feature requests

As a user I would like to have Isito installed on to my civo cluster

This issue is a:
[ ] Bug Report
[ ] Enhancement suggestion
[X] Feature request

Enhancement suggestions / feature requests

Would you consider adding a new category under Tools & Frameworks? I want to add RabbitMQ product but I couldn't choose a suitable category.

This issue is a: functionallity issue

check the attacked video to reproduce.

functionallity_issue.zip

This issue is a:
[ ] Bug Report
[ ] Enhancement suggestion
[x] Feature request

Enhancement suggestions / feature requests

In architecture section I think we should add nginx ingress controller. Currently it can be installed using helm install my-nginx stable/nginx-ingress. We just need to ensure that Traefik Ingress controller isn't selected during cluster creation. Otherwise we will get FailedScheduling issue for ingress controller pods because of ingress port unavailability.