helm: Error: failed to download "stable/nfs-server-provisioner"

hey @stefanthorpe! The nfs-server-provisioner chart is being downloaded from the stable repo. The first thing you’ll want to do is make sure that the helm repo list has the following line:

$ helm repo list
NAME  	URL
stable	https://kubernetes-charts.storage.googleapis.com
local 	http://127.0.0.1:8879/charts

If you do not see the stable repo listed, you can add it via $helm repo add stable https://kubernetes-charts.storage.googleapis.com

Can you try that out and let me know what you see?

Hi everyone

I faced the same issue on MacOS X.

Adding --debug helped me trace the problem which as file permissions/missing folder: Error: open /Users/user/.helm/cache/archive/prometheus-8.8.0.tgz: permission denied

I created the necessary folders and then it work. I also executed with sudo which also did the trick. Maybe this is not the case however i am justing adding here in case someone else hits the same issue like mine.

Hello Folks! I am taking the liberty to join the thread for I was about to open a new issue, but found this one first.

We have been experiencing a similar glitch, with the command helm upgrade with --install flag though. So the attempt was the following:

helm upgrade --install --namespace logmon --values ~/.kubernetes-yaml/logmon/prometheus-helm-values.yaml --version "v2.2.1" "prometheus" stable/prometheus

But getting the error message for older helm version:

Error: file “stable/prometheus” not found

Now in 2.91 updated to:

Error: failed to download “stable/prometheus” (hint: running helm repo update may help)

When I test both helm repo list and helm search prometheus, here are the outputs:

ubuntu@ip-10-0-90-213:~$ helm repo list
NAME            	URL                                             
stable          	https://kubernetes-charts.storage.googleapis.com
local           	http://127.0.0.1:8879/charts     

and

ubuntu@ip-10-0-90-213:~$ helm search prometheus
NAME                                 	CHART VERSION	APP VERSION	DESCRIPTION                                       
stable/prometheus                    	6.7.2        	2.2.1      	Prometheus is a monitoring system and time seri...
stable/prometheus-cloudwatch-exporter	0.1.2        	0.1.0      	A Helm chart for prometheus cloudwatch-exporter   
stable/prometheus-mysql-exporter     	0.1.0        	v0.10.0    	A Helm chart for prometheus mysql exporter with...
stable/prometheus-node-exporter      	0.2.0        	0.15.2     	A Helm chart for prometheus node-exporter         
stable/prometheus-postgres-exporter  	0.2.0        	0.4.6      	A Helm chart for prometheus postgres-exporter     
stable/prometheus-pushgateway        	0.1.2        	0.4.0      	A Helm chart for prometheus pushgateway           
stable/prometheus-rabbitmq-exporter  	0.1.3        	v0.28.0    	Rabbitmq metrics exporter for prometheus          
stable/prometheus-redis-exporter     	0.1.2        	0.16.0     	Prometheus exporter for Redis metrics             
stable/prometheus-to-sd              	0.1.0        	0.2.2      	Scrape metrics stored in prometheus format and ...
stable/elasticsearch-exporter        	0.1.4        	1.0.2      	Elasticsearch stats exporter for Prometheus       
stable/stackdriver-exporter          	0.0.4        	0.5.1      	Stackdriver exporter for Prometheus               
stable/weave-cloud                   	0.2.2        	0.2.0      	Weave Cloud is a add-on to Kubernetes which pro...
stable/kube-state-metrics            	0.7.1        	1.2.0      	Install kube-state-metrics to generate and expo...
stable/mariadb                       	4.2.2        	10.1.33    	Fast, reliable, scalable, and easy to use open-...

The head scratching session lasted for a while! Then together with another colleague we’ve noticed a discrepancy between the version we were passing as an argument and the actually correct one listed above (i.e 6.7.2 instead of 2.2.1 which is the app version, not the chart version) By changing the passed version to the correct one did help and now we can correctly install such chart.

Nevertheless we would like to propose to add as a return message not only the comment to try update the repo list, but to also double check that a valid existing chart version number is used.

Hope it could help and please let me know if this was actually obvious. IMHO, there is a difference between saying:

• Error: file “stable/prometheus” not found or Error: failed to download “stable/prometheus” (which is already somewhat more accurate)

and something like:

• Error: failed to download "stable/prometheus - We couldn’t found the specified chart version

or

• Error: failed to download "stable/prometheus - Argument X does not match current available chart’s attributes

In other words, a sort of case in-esac structure that will return a better output message.

Thanks and have a good one!

@stefanthorpe - wow that’s strange. What version of helm are you running? I’ll try to reproduce. Did you also try a helm repo update?

@carmat88 – thanks for your input. Agreed and it’s always good to have more detailed error messages! If you’d like to contribute and them more user friendly, please feel free.

I wonder if changing error messages like that breaks backwards compatibility. cc/ @technosophos