Tanzu Hub hubsm-install errand fails with "Get "https://<hub-url>/hub/graphql": dial tcp <ip>:443: connect: no route to host
search cancel

Tanzu Hub hubsm-install errand fails with "Get "https://<hub-url>/hub/graphql": dial tcp <ip>:443: connect: no route to host

book

Article ID: 437814

calendar_today

Updated On:

Products

VMware Tanzu Platform Core

Issue/Introduction

When attempting to install Tanzu Hub, it fails when running the hubsm-install errand like below:

==== 2026-04-22 02:34:06 UTC Running "/usr/local/bin/bosh --no-color --non-interactive --tty --environment=<env> --deployment=hub-<guid> run-errand hubsm-install --instance registry/first"
Using environment '<env> as client 'ops_manager'

Using deployment 'hub-<guid>'

...

           [i] Checking cron job status...  
           [i] Cron job ensemble-assessment-sync-repos-job is not suspended  
           [i] Cron job history-table-partition-maintenance is not suspended  
           [i] Cron job scheduled-backup is not suspended  
           [ok] Service health checks completed.  
           [i] Checking GraphQL API endpoint health...  
           [i] Checking GraphQL API endpoint accessibility...  
           [i] GraphQL API endpoint is: https://<url>/hub/graphql  
           [x] Installation failed with error: GET request to 'https://<url>/hub/graphql' failed: Get "https://<url>/hub/graphql": dial tcp <ip>:443: connect: no route to host  
           [i] Cluster logs collection start...  
           [i] Found 4 configmaps with label 'crash.tanzu.vmware.com/config'  
           [i] Stored crashd config successfully: /tmp/log-bundle/work/tanzusm/log-bundle-kube-capture/config.crash  
...

           [i] Successfully stored log bundle at /tmp/log-bundle/log-bundle-2026-04-22T06-35-38Z.tar.gz  
           [i] Log bundle execution took 3m5.608715589s  
           [x] Could not perform installation step: GET request to 'https://<url>/hub/graphql' failed: Get "https://<url>/hub/graphql": dial tcp <ip>443: connect: no route to host step="Verification checks will confirm the correctness of the installation and the health of the cluster."  
             
Stderr     using KUBECONFIG at /var/vcap/jobs/hubsm-install/config/kubeconfig  
           Capturing config: pod  
...

1 errand(s)

Environment

Tanzu Hub

Resolution

There are a few options to resolve this:

Option 1. Modify the hub domain record (the <url> in the above output) in DNS to point one of IP of 5 control VM.
Option 2: Assign <IP> to one of control VMs via bosh deploy command directly. Please note, this will be reverted on the next apply change, but the assigned IP will continue to be assigned.
Option 3: Configure your LB with the IP.

Tanzu Hub 10.4 will allow you to statically set control node IP's beforehand, just like the GoRouters, so you will be able to pre-create the DNS records.