SF Onboarding Week - January 27, 2020

[suprajanarasimhan]

Checklist

Communication

• Several weeks before: send out an email to the participants reminding them that they are signed up and to clear their calendars. You can find a template in FACILITATING.md.
  • You will be communicating with / authing these folks a lot. I find it helpful to save a copy/pasteable list of their names here:
    • Sara Arntz <sarntz@pivotal.io>, Judy van Soldt <jvansoldt@pivotal.io>, Samuel Waggoner <swaggoner@pivotal.io>
• The week before: send out an email to each of the participants. You can find a template in FACILITATING.md.
  - [ ] Get a Zoom room setup for remote participants. {} (Make sure to include this on your calendar invites.) . -- Added automated Zooms to calendar events instead.

Scheduling

• Schedule a welcome/orientation session (1hr at 9:30 Monday morning). Discuss the goals of the week and prime the participants to consider what they'd like to get out of Onboarding week.
• Schedule daily standup (15min at 9:15, starting Tuesday).
• Schedule afternoon check in (30min at 1:30).
• Schedule retro at the end of the week (1hr at 4:00), include link to retro board: https://retros.cfapps.io/retros/onboarding-sf (pw: onboarding-sf (pls don't change it))
• Schedule any sessions that you know the group would be interested in -- for example, BOSH boxes and lines.
  - [ ] In the past, non Engineers have appreciated it when we blocked out the entire days of onbaording. This is especially helpful if done a few weeks in advance; so, they have time to move meetings and avoid having new ones scheduled in the first place. -- Great idea, but missed this time.

---

These are core CF topics, other topics are a bit more "extra credit".

• BOSH {}
• CF Overview {}
  - [ ] Concourse {} -- Added a k8s talk instead.

As a helpful reminder, you can save the name of the speaker w/ checkbox.

IaaS Setup

• Make sure you have GCP projects set up.
  • In SF, we use:
    • cf-sf-onboarding-env-1
    • cf-sf-onboarding-env-2
    • cf-sf-onboarding-env-3
    • cf-sf-onboarding-env-4
    • cf-sf-onboarding-env-5
      ~- [ ]~Otherwise, Fill out this form to create new environments. -- New environments did not need to be created.
• Clean up any VMs, networks, and DNS records that may have been leftover from the previous week
  • Take a look at this bbl command
• Add all participants as owners to each of the GCP projects.
  • In SF:
    • IAM for env-1
    • IAM for env-2
    • IAM for env-3
    • IAM for env-4
    • IAM for env-5
• If your CF on-boarding will include the GCP track (gcp tracker label / the oss build) then ensure participants have access to the CF-Onboarding-dns project in GCP (for DNS propagation). IAM for DNS

Tracker Setup

• Make sure you have Tracker projects prepared.
  • In SF, we use the following trackers projects:
    • Pair 1
    • Pair 2
    • Pair 3
    • Pair 4
    • Pair 5
• Add all participants as members to each of the tracker projects. (A list of emails works here.)
• Delete everything from the tracker projects.
• Build the tracker back log csv
  • Run ./build oss from the onboarding project, this will generate onboarding-tracker.csv
    • Note: you need docker installed brew cask install docker
      • To get the command line tool to show up you may have to first start docker via Spotlight Search
      • If docker gives you this error: failed to dial gRPC: unable to upgrade to h2c, received 502 context canceled
        • In docker settings: enable experimental, restart docker, disable experiment and restart again.
• Import the CSV into each Tracker project (Taskbar > More > Import CSV > Choose File)

LastPass Setup

• Give all participants access to the shared LastPass folder:
  • LastPass > Sharing Center > Shared-CF SF Onboarding > Manage > Invite Users (A list of emails works here.)

Workstation Setup

Note: local IT departments can sometimes help with workstation setup. (Helpful for remote facilitation.)

• Make sure you have a workstation for each pair.
• Ensure ethernet cable is plugged into iMac, not monitor.
• Re-image each machine. Here are steps:
• Open a terminal window and run

curl -LOk https://github.com/pivotal-cf-experimental/onboarding-week-construct/archive/master.zip
unzip master.zip
cd onboarding-week-construct-master
./construct

You'll need to babysit this process, as it will prompt you for the password several times.

Misc

• Record any issues you have with this process so we can update this checklist.

• Imaging server was inaccessible. Got help from Kevin on the OSM team to re-image the machine's using their USB copy of Catalina. Need to update the WORKSTATION_IMAGING.md document to indicate this and explain how to re-image this alternate way in case the imaging server is down in the future.
• Feedback from Emily B. - Docker part was confusing. Not clear that you have to click on the Docker whale icon to start the Docker daemon.
• When running ./construct to install all tools on the Onboarding workstation:
  -- Problem:
  Ran into a problem seen during the November 2019 Onboarding workstation setup:

-* installing homebrew-casks

Error: caskroom/cask was moved. Tap homebrew/cask-cask instead.

-- Workaround solution:
Type brew cask, which seems to install the brew cask tool. Then delete the directory onboarding-week-construct/homebrew-casks

==> Tapping homebrew/cask
Cloning into '/usr/local/Homebrew/Library/Taps/homebrew/homebrew-cask'...
...
Homebrew Cask provides a friendly CLI workflow for the administration
of macOS applications distributed as binaries.

Commands:

    --cache    display the file used to cache the Cask
...

-- Ideal solution:
In https://github.com/pivotal-cf-experimental/onboarding-week-construct repository, change homebrew-casks/install.sh from:

brew tap cask/cask

to:

brew cask

However this fix results in the need to run ./construct multiple times:
-- The first time it fails with:

>>> Successfully installed ruby 2.4.0 into /Users/pivotal/.rubies/ruby-2.4.0
zsh:1: command not found: chruby

-- The second time it fails with:

ln: /usr/local/bin/vi: File exists

-- The third time it succeeds.

Guess why: Perhaps there is need for changing the order of installation of some of these tools?

-- References:
Found this solution in this helpful post.

Workstation Setup Action Items

• Update the WORKSTATION_IMAGING.md document to indicate alternate way to re-image when imaging server is down.
• Follow-up on Docker instructions via this branch - Remove workstation imaging script changes so only Docker changes are on this branch.
• Follow-up on ./construct revisions via this fork. -- Ask to be added as a contributor via ask@ ticket.

[suprajanarasimhan]

We can consider this checklist complete with the exception of the "Workstation Setup Action Items."