Installing on GCP. Node not found

gravity
#1

Hi,
I’m trying to install the mattermost.tar that I got from the quickstart tutorial. This installation works on an Ubuntu VM on my laptop. However I tried to install it on a VM on GCP and it failed. My internal IP is 10.128.0.3 and the name of the GCP VM is on-prem-host. As you can see below, I call the install with 10.128.0.3.

[ERROR]: planet is not running yet: &{degraded [{ 10.128.0.3 master  degraded [nodestatus (node "on-prem-host" not found)]}]}, failed to execute phase "/health" (pl
anet is not running yet: &{degraded [{ 10.128.0.3 master  degraded [nodestatus (node "on-prem-host" not found)]}]})

Full Details below:

rajiv_abraham1@on-prem-host:~$ sudo ./gravity install --advertise-addr=10.128.0.3 --token=a_password
Thu Dec  5 20:54:01 UTC Starting installer
..Thu Dec  5 20:54:13 UTC Successfully added "node" node on 10.128.0.3
Thu Dec  5 20:54:13 UTC All agents have connected!
Thu Dec  5 20:54:14 UTC Operation has been created
Thu Dec  5 20:54:15 UTC Executing "/checks" locally
Thu Dec  5 20:54:15 UTC Execute preflight checks
Thu Dec  5 20:54:15 UTC Running pre-flight checks
Thu Dec  5 20:54:18 UTC Executing "/configure" locally
Thu Dec  5 20:54:19 UTC Configuring cluster packages
Thu Dec  5 20:54:19 UTC Configure packages for all nodes
Thu Dec  5 20:54:22 UTC Executing "/bootstrap/on-prem-host" locally
Thu Dec  5 20:54:23 UTC Bootstrap master node on-prem-host
Thu Dec  5 20:54:23 UTC Configuring system directories
Thu Dec  5 20:54:26 UTC Configuring application-specific volumes
Thu Dec  5 20:54:27 UTC Pull configured packages
Thu Dec  5 20:54:27 UTC Executing "/pull/on-prem-host" locally
Thu Dec  5 20:54:27 UTC Pulling user application
Thu Dec  5 20:54:28 UTC Pull packages on master node on-prem-host
Thu Dec  5 20:54:37 UTC         Still pulling user application (10 seconds elapsed)
Thu Dec  5 20:54:42 UTC Pulling configured packages
Thu Dec  5 20:54:44 UTC Unpacking pulled packages
Thu Dec  5 20:54:54 UTC         Still unpacking pulled packages (10 seconds elapsed)
Thu Dec  5 20:55:04 UTC         Still unpacking pulled packages (20 seconds elapsed)
Thu Dec  5 20:55:05 UTC Install system software on master nodes
Thu Dec  5 20:55:05 UTC Executing "/masters/on-prem-host/teleport" locally
Thu Dec  5 20:55:06 UTC Installing system service teleport:3.2.13
Thu Dec  5 20:55:06 UTC Install system package teleport:3.2.13 on master node on-prem-host
Thu Dec  5 20:55:08 UTC Executing "/masters/on-prem-host/planet" locally
Thu Dec  5 20:55:08 UTC Install system package planet:6.1.9-11505 on master node on-prem-host
Thu Dec  5 20:55:08 UTC Installing system service planet:6.1.9-11505
Thu Dec  5 20:55:18 UTC         Still installing system service planet:6.1.9-11505 (10 seconds elapsed)
Thu Dec  5 20:55:28 UTC         Still installing system service planet:6.1.9-11505 (20 seconds elapsed)
Thu Dec  5 20:55:30 UTC Executing "/wait" locally

Thu Dec  5 20:55:31 UTC Wait for Kubernetes to become available
Thu Dec  5 20:55:40 UTC         Still executing "/wait" locally (10 seconds elapsed)
Thu Dec  5 20:55:47 UTC Executing "/rbac" locally
Thu Dec  5 20:55:47 UTC Creating Kubernetes RBAC resources
Thu Dec  5 20:55:48 UTC Bootstrap Kubernetes roles and PSPs
Thu Dec  5 20:55:50 UTC Executing "/coredns" locally
Thu Dec  5 20:55:50 UTC Configure CoreDNS
Thu Dec  5 20:55:50 UTC Configuring CoreDNS
Thu Dec  5 20:55:51 UTC Executing "/system-resources" locally
Thu Dec  5 20:55:51 UTC Create system Kubernetes resources
Thu Dec  5 20:55:51 UTC Configuring system Kubernetes resources
Thu Dec  5 20:55:52 UTC Executing "/export/on-prem-host" locally
Thu Dec  5 20:55:53 UTC Populate Docker registry on master node on-prem-host
Thu Dec  5 20:55:53 UTC Unpacking application rbac-app:6.1.13
Thu Dec  5 20:55:53 UTC Exporting application rbac-app:6.1.13 to local registry
Thu Dec  5 20:55:53 UTC Unpacking application dns-app:0.3.0
Thu Dec  5 20:55:54 UTC Exporting application dns-app:0.3.0 to local registry
Thu Dec  5 20:55:56 UTC Unpacking application bandwagon:6.0.1
Thu Dec  5 20:55:56 UTC Exporting application bandwagon:6.0.1 to local registry
Thu Dec  5 20:55:58 UTC Unpacking application logging-app:6.0.2
Thu Dec  5 20:55:59 UTC Exporting application logging-app:6.0.2 to local registry
Thu Dec  5 20:56:03 UTC Unpacking application monitoring-app:6.0.4
Thu Dec  5 20:56:05 UTC Exporting application monitoring-app:6.0.4 to local registry
Thu Dec  5 20:56:14 UTC Unpacking application tiller-app:6.1.0
Thu Dec  5 20:56:14 UTC Exporting application tiller-app:6.1.0 to local registry
Thu Dec  5 20:56:15 UTC Unpacking application site:6.1.13
Thu Dec  5 20:56:16 UTC Exporting application site:6.1.13 to local registry
Thu Dec  5 20:56:18 UTC Unpacking application mattermost:2.2.0
Thu Dec  5 20:56:21 UTC Exporting application mattermost:2.2.0 to local registry
Thu Dec  5 20:56:30 UTC Executing "/health" locally
hu Dec  5 20:56:30 UTC Waiting for the planet to start
Thu Dec  5 20:56:31 UTC Wait for cluster to pass health checks
Thu Dec  5 20:56:40 UTC         Still waiting for the planet to start (10 seconds elapsed)
...
hu Dec  5 21:04:50 UTC         Still waiting for the planet to start (8 minutes elapsed)
Thu Dec  5 21:04:52 UTC Executing operation finished in 10 minutes
Thu Dec  5 21:04:52 UTC Saving debug report to /home/rajiv_abraham1/crashreport.tgz
[ERROR]: planet is not running yet: &{degraded [{ 10.128.0.3 master  degraded [nodestatus (node "on-prem-host" not found)]}]}, failed to execute phase "/health" (pl
anet is not running yet: &{degraded [{ 10.128.0.3 master  degraded [nodestatus (node "on-prem-host" not found)]}]})
rajiv_abraham1@on-prem-host:~$
#2

I found the doc. I’ll execute the instructions. https://gravitational.com/gravity/docs/installation/#google-compute-engine

#3

When running gravity on a cloud provider like GCE, it’s going to try and turn on the cloud provider integrations by default. Which the doc you linked to describes.

Alternatively if you just want to test how gravity will operate on a GCE VM as if it were on-prem, there is an installer flag to force the integrations into on prem mode. You would need to pass the --cloud-provider=generic flag when starting the install.

The flag is documented here: https://gravitational.com/gravity/docs/installation/#cli-installation

#4

Thank you. That’s valuable to know.

#5

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.