[BUG] harvester-installer: errors in validateConfig, PrepareWebooks and doInstall appear in console, but aren't logged

[tserong]

Describe the bug The last chunk of code in the install panel's PreShow() function calls validateConfig(), PrepareWebhooks() and one of configureInstalledNode() or doInstall(). All these functions return an error value in err, but if err is set, all we do is call printToPanel() - we never actually log the error:

https://github.com/harvester/harvester-installer/blob/36ea2bf0c7f4e420b5ca4f1d1a75312240d1f9bc/pkg/console/install_panels.go#L2168-L2184

This means if you're using the installer interactively and one of these things fails, you can see what went wrong on the console display. But, if you're doing a PXE install and any of these things fail, and for some reason you can't see the console and can only access the log in /var/log/console.log you will have no idea what went wrong.

To Reproduce Steps to reproduce the behavior:

1. Set up a PXE install (e.g. vagrant-pxe-harvester from https://github.com/harvester/ipxe-examples), but mess up the config somehow, for example:

   install:
    mode: This ain't right

2. Boot a node, and confirm that the installation fails by checking the console:
3. Check /var/log/console.log and note that there's no indication of the actual problem - in this example all we see is some preflight check failures which is quite misleading:

Expected behavior Whatever failed is actually logged.

Support bundle N/A

Environment

• Harvester ISO version: v1.3.1
• Underlying Infrastructure (e.g. Baremetal with Dell PowerEdge R630): virtualized

Additional context N/A

[harvesterhci-io-github-bot]

added backport-needed/1.3.2 issue: #6215.

[harvesterhci-io-github-bot]

Pre Ready-For-Testing Checklist

• [x] Where is the reproduce steps/test steps documented? The reproduce steps/test steps are at:

  • In the description of this issue
  • Also in the associated PR (https://github.com/harvester/harvester-installer/pull/775)

• [x] Is there a workaround for the issue? If so, where is it documented? The workaround is at: Fiddle with kernel boot parameters until you get the console visible somewhere and are able to see the errors (e.g. try console=tty1 instead of console=ttyS0 if possible)

• [x] Have the backend code been merged (harvester, harvester-installer, etc) (including backport-needed/*)? The PR is at: https://github.com/harvester/harvester-installer/pull/775

  • [x] Does the PR include the explanation for the fix or the feature?

  • [ ] Does the PR include deployment change (YAML/Chart)? If so, where are the PRs for both YAML file and Chart? The PR for the YAML change is at: The PR for the chart change is at:

• [ ] If NOT labeled: not-require/test-plan Has the e2e test plan been merged? Have QAs agreed on the automation test case? If only test case skeleton w/o implementation, have you created an implementation issue?
  • The automation skeleton PR is at:
  • The automation test case PR is at:

[harvesterhci-io-github-bot]

Automation e2e test issue: harvester/tests#1401

[irishgordo]

@tserong thanks for this enhancment :smile: :+1: - this all looks good - I'll go ahead and close this out :smile:

Environment

• Harvester
  • Version: Harvester-v1.4-head
  • Profile: QEMU/KVM
  • ui-source: Auto

Verification

1 Node Cluster

1. :green_circle: Logs are more useful given strange parameters in Harvester configuration at Install time w/ Harvester Installer

   Given a strange harvester install mode is specified

   • our installer should be logging out the content of what has happened giving us more depth in the console about the yaml configuration it was provided