Good afternoon I tried to make a backup of the EKS cluster u

Question

Good afternoon I tried to make a backup of the EKS cluster using Velero <https velero io > It turns out that when restoring the backup all containers went up normally except for Orchest Could you help me understand I m trying to implement a backup routine here at the Dadosfera company Orchest Controller has failed

Rick Lamers · Answer

Thanks for reporting we ll investigate what could be the issue

Alexsander Pereira · Answer

Thank you I ll wait

Rick Lamers · Answer

We re still investigating this issue we expect to have an update tomorrow Friday

Yannick · Answer

Thank you for the error logs < Alexsander Pereira> pray < Navid H> Could you take a look at Thank you

Navid H · Answer

< Alexsander Pereira> I ll look into this and will comeback as soon as possible

Yannick · Answer

< Alexsander Pereira> FYI a fix will be included in the next release thanks to < Navid H> star struck

Yannick · Answer

Released nerd face

Alexsander Pereira · Answer

< Yannick> That s great news I ll do the tests

Alexsander Pereira · Answer

I upgraded to the new version and this is what happened The orchest webserver and orchest api did not go up Can you help me

Rick Lamers · Answer

We ll check out the logs and get back to you tomorrow < Alexsander Pereira> At first glance it looks like it s finding that resources already exist in the cluster that the controller doesn t expect to find Could you describe the actions you took to get to the error log you posted

Yannick · Answer

< Navid H> With the addition of deploying Nginx as part of Orchest do you think this could have caused any of the issues One of the logs that made me think this is the case ```E0914 17 51 53 278362 1 controller go 172 failed to deploy helm deployment Error rendered manifests contain a resource that already exists Unable to continue with install IngressClass nginx in namespace exists and cannot be imported into the current release invalid ownership metadata label validation error missing key must be set to Helm annotation validation error missing key must be set to orchest ingress nginx annotation validation error missing key must be set to orchest ``` gt I upgraded to the new version and this is what happened < Alexsander Pereira> Did you try Velero again or did this happen just by doing a regular `orchest update`

Navid H · Answer

Hi < Alexsander Pereira> Could you run the following command and return the result back to us Thanks ```kubectl get ingressclass```

Alexsander Pereira · Answer

```alex alex Legion 5 Pro 16ITH6 ~ Documents code dadosfera$ kubectl get ingressclass NAME CONTROLLER PARAMETERS AGE nginx <http k8s io ingress nginx|k8s io ingress nginx> lt none gt 39h```

Alexsander Pereira · Answer

< Yannick> < Navid H> I install Nginx just like it worked before Was that why Our nginx is customized with ACM certificate and tags on the load balancer I need to keep having control over it Any way to disable nginx installation in orchest controller

Alexsander Pereira · Answer

Or would it be possible to customize this in the orchest controller and not install nginx separately anymore

Navid H · Answer

< Alexsander Pereira> Thanks for getting back to us This is a corner case that we will fix in the next release but in the mean time you can add the following annotation to the OrchestCluster to disable Nginx ``` false ``` After that you should also remove the nginx application from your `OrchestCluster` CRD

Navid H · Answer

< Alexsander Pereira> Also one more question are you using eks or gke cluster

Yannick · Answer

gt I tried to make a backup of the EKS cluster using Velero point up So I guess EKS

Alexsander Pereira · Answer

It worked Very good < Yannick> < Navid H> Thanks

Alexsander Pereira · Answer

< Navid H> We use EKS

Navid H · Answer

< Alexsander Pereira> Thanks for letting us know pray