Thank you @A_smart_kitten! Updated task to indicate that has been determined. Will get the change reviewed to ensure nothing breaks in release versions before or after (although unlikely).

Figure out what version we were on when we set $wgUseCodexSpecialBlock and $wgEnableMultiBlocks to universally true

Thank you @Dzahn! :)

Hello @Dzahn, got your Email and sent over a verification response. Thanks for getting to that so swiftly :))

Actually, might also be the case that my other username on here (accidentally created when I was no longer a contractor) is the issue. The user was EBomani-WMF and account has since been disabled, though, so I'm not sure.

Thank you so much, @thcipriani, @Dzahn and @Aklapper!

I joined as a contractor but converted to full-time about a year ago. I am not sure why it still indicates that and not sure what to do to fix it. Please inform and I will get to it.

Completed during office hours today. Leaving "In Progress" to mark resolved during planning.

The bot is now reporting appropriately for all created and deleted environments 🎉.

Updated the mockup to reflect the agreed upon design. Please see images below.

Attaching the images from the latest design discussion we had. Our main objectives were surrounding usability and reducing visual clutter wherever possible.

It came to my attention that the previous images are not showing up for others despite looking good on my end. This is my attempt to fix this, hopefully that works.

Created some mocks for what we could do on the front-end side to represent the Wikifarms on Patch demo. The current version has it such that :

• If 'CampaignEvents' extensions is clicked, the option to create as a wikifarm becomes enabled and so does the text box asking for how many of them to create (probably should do a drop down and have an upper limit for this)
• Once decided, the creation process happens as it would normally but the screen lasts longer as it would need to complete all wikifarm wikis.
• When done, the wikifarm created will be displayed on the table similarly to how other wikis are except there will be a clickable option that exposes all the wikis that are contained in a wikifarm.

The bot now announces creation and deletion of both legacy and catalyst backend environments created. My most recent tests were done on staging with the merged API changes and the creation and deletion logs can be seen on the test task.

This was fixed and merged. Staging is now back to normal.

The UI was affected by recent changes that converted the wikis table into a form which inadvertently made the form rule on max-width also apply to the main wiki "table" (in quotes because it is now a form not a table by definition) . This limited width should not be applied to the wiki "table". The display looked as follows:

PatchDemo! Sorry for the unnecessary ping, will delete soon :)

From my investigation into this, I found the following about potentially using OpenTofu for infrastructure and configuration management on Catalyst:

For more context regarding our decision to not use codex for this change, here are some notes from my talks with Codex folks :

Implemented and merged. Announcement up :)

I reached to @KColeman-WMF and got some very enlightening and useful UX and Design direction for this update. From that, we have decided to implement the version with the checkboxes visible to users by default and will no longer need to send the survey as the initial reasons for omitting the checkboxes have been adequately and beautifully addressed in the design she very kindly provided us to use as a guideline and the hidden checkboxes version is rightfully no longer an option we can recommend. I am currently working on testing out whether we can successfully implement this using Codex inspired CSS elements for the table without necessarily needing a complete overhaul of the Patchdemo UI. Will update the task with the final visual representation of the prototype for approval as soon as I can get that looking good and working well.

After our discussion during the planning, these were the viable workflows for this process that we decided to bring forth and get opinions on.

Working on this task. I narrowed down the most intuitive/effective UI process as follows:

The SSH issue seems to have been sorted. We are uncertain what exactly happened but here are the two ways the instances were able to be SSH'd into.

In the process of troubleshooting using a config file to launch instances. Currently this results in an inability to SSH onto the instance and it is likely due to some processes outlined in the config that we need to modify/update.

@Aklapper done, thank you!

Synced up with @Mhurd and confirmed that the OpenTofu scripting worked exceptionally and functioned as expected even when used with the scripts🥳. We leveraged this to create a new ci-components instance and will be monitoring and adjusting it as needed.

From my conversations with @Mhurd, it was determined that although it was not explicitly noted on this task, it is important to figure out how to make the OpenTofu resource-creation for instances work with configuration scripts as they are integral to the respective instances' functionality. Due to that, I have done further investigation and believe that it is in fact possible to do so and will be meeting with Monte later on this week to confirm my findings and officially test out creating identical copies of either ci-selenium-components/ci2 instances.

@Jdlrobson-WMF ReadingLists extension is now live and selectable on Patchdemo production. Installation on wikis is also verified via the Special:Version page. Thanks.

The current configuration is capable of creating new instances and volumes and linking them to one another as well as dynamically updating the volume size whenever needed. For the purposes of the task, the following is done:

@dcaro Thank you! Since we already used to get emails for Puppet failures on all our instances and it seems that manually done InstanceDown events do not trigger active alerts for e-mails, I think the only way we can assert this works is when the instances experience issues organically in the future. I am happy to close out this issue for now and if we do notice an issue that occurs but does not provide e-mail alerts, we can reopen it. Thanks so much again for your help with this!