Based on https://github.com/adamhgriffith-uofu/dualstack-virtualbox-k8s update the documentation for a manual cluster installation using kubeadm.

Write a blog post for the upcoming SLATE API & Client v1.0.33 combo. Rough release notes include:

• Rocky 9
• AWS C++ SDK 1.9.365
• OpenTelemetry 1.6.1
• Latest NGINX ingress

https://slateci.io/technology/ has mentions of Singularity as an alternative container runtime to Docker, but singularity-cri has been deprecated: https://github.com/sylabs/singularity-cri

Use the following resources:

• Slack conversations
• SLATE Upgrade Guide .. Guide
• SLATE Kubernetes Migration Guide: 1

When signing up, a new user should be shown https://github.com/slateci/slateci.github.io/blob/master/_docs/security-and-policies/acceptable-use.md and required to indicate assent before the account is created/finalized.

We need to clearly mark how the project is funded, i.e. by the NSF and the grant number. https://www.nsf.gov/awardsearch/showAward?AWD_ID=1724821 should have the appropriate details. This information can probably go in the front page footer.

Add an explicit dependabot.yml file to this repository.

Getting the following error:

======================================================================
FAIL: test_docs_page (__main__.WebPageBrowsing)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/home/runner/work/slateci.github.io/slateci.github.io/slateci-unit-tests/main.py", line 155, in test_docs_page
    docs_page.iterate_links_doc_content(linkL1, linkL2)
  File "/home/runner/work/slateci.github.io/slateci.github.io/slateci-unit-tests/page.py", line 154, in iterate_links_doc_content
    assert self.is_page_valid()
AssertionError

----------------------------------------------------------------------
Ran 6 tests in 206.988s

FAILED (failures=1)

The current infrastructure is not yet ready for customers to adopt dual-stack. Modify the SLATE Manual Cluster Installation guide to once again reflect a single-stack IPv4 process.

There is an issue with docs pages jumping when you scroll the page. This only happens when the window is minimized to the point where the menus become hamburgers.

This is a suggestion to add a headings-based TOC to the blog HTML template.

Steps to repeat:

1. Navigate to https://slateci.io/docs/security-and-policies/index.html.
2. Click Security and Policies in the left navigation.
3. Scroll to the bottom of the page.
4. Note that Resources in the left navigation is now covered by the footer.

Related Selenium failure text:

======================================================================
ERROR: test_docs_page (__main__.WebPageBrowsing)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/opt/project/main.py", line 103, in test_docs_page
    side_menu_btns[i].click()
  File "/usr/local/lib/python3.9/site-packages/selenium/webdriver/remote/webelement.py", line 80, in click
    self._execute(Command.CLICK_ELEMENT)
  File "/usr/local/lib/python3.9/site-packages/selenium/webdriver/remote/webelement.py", line 633, in _execute
    return self._parent.execute(command, params)
  File "/usr/local/lib/python3.9/site-packages/selenium/webdriver/remote/webdriver.py", line 321, in execute
    self.error_handler.check_response(response)
  File "/usr/local/lib/python3.9/site-packages/selenium/webdriver/remote/errorhandler.py", line 242, in check_response
    raise exception_class(message, screen, stacktrace)
selenium.common.exceptions.ElementClickInterceptedException: Message: element click intercepted: Element <a href="/docs/resources/k8s-certificates.html" role="button" aria-expanded="false" aria-controls="resources" class="doc-nav-top-link">...</a> is not clickable at point (48, 912). Other element would receive the click: <footer id="footer">...</footer>
  (Session info: headless chrome=91.0.4472.101)


----------------------------------------------------------------------

Requested by Rob

This should be integrated into the application development pages.

Update Obtain a SLATE Token to refer to the Portal CLI Access as much as possible to reduce duplicated information.

• The current blog source
• Change the date to today.
• Move out of drafts

Installing a fresh cluster today with MetalLB v0.10.2 as per the instructions, I ran into a problem that required me to do the following:

kubectl create secret generic -n metallb-system memberlist --from-literal=secretkey="$(openssl rand -base64 128)"

I am not sure if this is a new, general problem with MetalLB, or a problem between keyboard and chair.

Perhaps we should update the instructions for MetalLB v0.11.0, as is recommended by upstream? https://metallb.universe.tf/installation/

It may obviate this issue.

Homogenize inputs, copy boxes, etc.

The footer has a copyright from 2017 to 2020. It may be good to update that to 2017-2022.

Update the Metallb documentation on the manual cluster install documents.

Sometimes we encounter timeout errors in the Integration tests (e.g. from Add example logs, fix a typo (#184) #107:

======================================================================
ERROR: test_comm_page (__main__.WebPageBrowsing)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/home/runner/work/slateci.github.io/slateci.github.io/slateci-unit-tests/main.py", line 210, in test_comm_page
    comm_page.wait_for_page_loaded()
  File "/home/runner/work/slateci.github.io/slateci.github.io/slateci-unit-tests/page.py", line 211, in wait_for_page_loaded
    WebDriverWait(self.driver, 20).until(
  File "/opt/hostedtoolcache/Python/3.9.7/x64/lib/python3.9/site-packages/selenium/webdriver/support/wait.py", line 90, in until
    raise TimeoutException(message, screen, stacktrace)
selenium.common.exceptions.TimeoutException: Message: 
Stacktrace:
#0 0x5585a80722c3 <unknown>
#1 0x5585a7e7b83a <unknown>
...
...

I believe this is because the Integration test workflow is running before the GitHub pages workflow is done publishing the new site content. Both work workflows trigger from push events to master so let's try a workflow_run event for the Integration Tests (see workflow_run for more details.

selenium.common.exceptions.SessionNotCreatedException: Message: session not created: This version of ChromeDriver only supports Chrome version 92 Current browser version is 96.0.4664.110 with binary path /usr/bin/google-chrome

Example of error:

• slateci.io Selenium Tests #180

Places to start:

• Update the ChromeDriver

https://slateci.io/docs/cluster/

FYI, none of the links to hardware work. I tried to take a stab at making this work again but couldn't figure it out. Could you take a look @Spelkington ?

As per @LincolnBryant we should remove references on our website to the now offline SLATE sandbox.

today I ran into a few helm issues:

Error: 1 chart(s) linted, 1 chart(s) failed
	%!s(<nil>)
	[INFO] Chart.yaml: icon is recommended
	[ERROR] templates/: render error at (condor-login/templates/deployment.yaml:109): unexpected EOF

This was due to a missing {{ end }}

Likewise I also ran into a problem with _helpers.tpl, which ultimately ended up being:

diff --git a/incubator/condor-login/condor-login/templates/pvc.yaml b/incubator/condor-login/condor-login/templates/pvc.yaml
index 08ce899..d840ab1 100644
--- a/incubator/condor-login/condor-login/templates/pvc.yaml
+++ b/incubator/condor-login/condor-login/templates/pvc.yaml
@@ -1,9 +1,9 @@
 apiVersion: v1
 kind: PersistentVolumeClaim
 metadata:
-  name: {{ template "condor-login.name" . }}-config-osg
+  name: {{ template "condor-login.fullname" . }}-config-osg
   labels:
-    app: {{ template "condor-login.name" . }}
+    app: {{ template "condor-login.fullname" . }}
     chart: {{ template "condor-login.chart" . }}
     release: {{ .Release.Name }}
     instance: {{ .Values.Instance }}
@@ -18,9 +18,9 @@ spec:
 apiVersion: v1
 kind: PersistentVolumeClaim
 metadata:
-  name: {{ template "condor-login.name" . }}-connect
+  name: {{ template "condor-login.fullname" . }}-connect
   labels:
-    app: {{ template "condor-login.name" . }}
+    app: {{ template "condor-login.fullname" . }}
     chart: {{ template "condor-login.chart" . }}
     release: {{ .Release.Name }}
     instance: {{ .Values.Instance }}

Note the lack of . in the previous versions.

Other stuff like that we need to document..

This should go in _docs/concepts/index.md

After upgrading from K8s v1.23.x to v1.24.x an admin will need to remove --network-plugin from /var/lib/kubelet/kubeadm-flags.env. Otherwise the kubelet will not restart.

While dockerd is headed out the door the following is necessary when navigating the current manual SLATE installation instructions.

mkdir /etc/docker
cat <<EOF | sudo tee /etc/docker/daemon.json
{
  "exec-opts": ["native.cgroupdriver=systemd"],
  "log-driver": "json-file",
  "log-opts": {
    "max-size": "100m"
  },
  "storage-driver": "overlay2"
}
EOF

Resources:

• Container Runtimes: Docker

This should have details about the client/server, the security model, etc.

We need to first gather logos/icons for applications we support, then we should show some of them on the front page to get people interested in what they could run.

Background

The SLATE catalog is moving to the following GitHub repos:

• https://github.com/slateci/slate-catalog-incubator
• https://github.com/slateci/slate-catalog-stable
  where in-house Dockerfiles and related materials will be stored and released by https://github.com/slateci/docker-images.

Task

Update the external-facing material at https://slateci.io/docs/apps to reflect this new information. Once complete create a pull request and submit to the team for review.

As pointed out by @fengpinghu, incorporate the Upgrading an installation that uses manifests and the Kubernetes API datastore into the blog post.

Add something like:

concurrency:
  group: integ-tests-${{ github.repository }}-${{ github.ref }}

to .github/workflows/integration-tests.yml.

The instructions for setting up MetalLB on OpenStack (55ba3c8) appear to have been removed (unless I am overlooking them). These are important to include, as this is a case we use fairly frequently ourselves and is often requests by sites who are interested in trying out SLATE.

The Automated Cluster Information guide needs an update to reflect lessons learned by new students and staff.

Verify that the cluster data on the Google Map is accurate, and that cluster information isn't being used in the headers of nodes.

It needs more details.