Error: UPGRADE FAILED: "review-feature-te-f5jdha" has no deployed releases

I have got our gitlab installation integrated with our kubernetes cluster, and setup a test project using autodeploy. When the review stage came around, it all worked fine. However, I am now trying to move a real project to use this workflow, and on the exact same review stage, I see the error:

 $ auto-deploy deploy
 secret "review-feature-te-f5jdha-secret" deleted
 secret/review-feature-te-f5jdha-secret replaced
 Using helm values file '.gitlab/auto-deploy-values.yaml'
 Deploying new stable release...
 Error: UPGRADE FAILED: "review-feature-te-f5jdha" has no deployed releases
 Error: "review-feature-te-f5jdha" has no deployed releases

I installed helm/tiller view the UI in gitlab, along with the ingress and prometheus - all of which seem to work fine on my demo project. I’ve seen other issues related to this, and people recommended using the helm cli tool, so I install that (after setting up the TLS stuff to connect to the tiller instance created by gitlab), I see none of my apps in the releases and no was of purging them. Is there something I am missing here?

$ helm ls --tiller-connection-timeout 30 --tls --tls-ca-cert tiller-ca.crt --tls-cert tiller.crt --tls-key tiller.key --all
NAME       	REVISION	UPDATED                 	STATUS  	CHART               	APP VERSION	NAMESPACE          
certmanager	1       	Wed Apr 22 11:08:10 2020	DEPLOYED	cert-manager-v0.10.1	v0.10.1    	gitlab-managed-apps
ingress    	1       	Wed Apr 22 10:04:56 2020	DEPLOYED	nginx-ingress-1.29.7	0.28.0     	gitlab-managed-apps
prometheus 	1       	Wed Apr 22 11:26:37 2020	DEPLOYED	prometheus-9.5.2    	2.13.1     	gitlab-managed-apps
runner     	1       	Wed Apr 22 11:09:11 2020	DEPLOYED	gitlab-runner-0.14.0	12.8.0     	gitlab-managed-apps

I am running self-hosted GitLab12.9.4-ee (6a1a8e88568)

I don’t have a gitlab-ci.yml, but have disabled some of the stages using variables (the same stages skipped by the demo project I created)

Any help or advice would be great at this point

I think I may have found the issue for anyone looking. I think it’s because the kubernetes deployment was failing do to secrets being missing. I’m now getting a different issue, but it seems tiller fails to find the deployment if the kube deployment fails