#random (2020-01)
Non-work banter and water cooler conversation
A place for non-work-related flimflam, faffing, hodge-podge or jibber-jabber you’d prefer to keep out of more focused work-related channels.
Archive: https://archive.sweetops.com/random/
2020-01-02
Also pretty nice find: https://github.com/oceanicsdotio/bathysphere
Has someone used Keel before? It seems to be a nice tool to manage Helm Deployments as well. https://keel.sh/
2020-01-03
(j/k, I heard of it but haven’t tinkered yet)
I just like the automated update of stuff.. so it will check if there is a new version and just update.
2020-01-05
Pretty neat implementation that stores all email on s3 and uses plus addressing for folders
2020-01-06
2020-01-08
lololol the evil part of me wants to feed that to git newbies
I was thinking the exact same thing. Copy/paste the output and ask why they don’t know abou the ‘git-lick-ref’ git subcommand already….
This message was deleted.
I try to only click on links sent by random people but that article set me straight.
lol
2020-01-09
Any suggestions about code coverage tools ?
Sonarqube
thanks
I heard that sonarqube is pretty complex to setup
Nice
Thanks
2020-01-10
2 most underrated kubernetes tools: krew (plugin manager adding kubectl commands you never knew you needed). jx (jenkins-x) - Even if you don’t use this for pipelining apps it has so many freaking features and tooling (like spitting out terraform for a new cluster in different cloud providers) that its hard not to like.
happy friday all
spitting out terraform for a new cluster in different cloud providers
Actually, it seems like it only creates the terraform for gke ATM, how sad (and that’s what I get for making assumptions).
Def not using that TF for my prod CI, thanks.
2020-01-11
2020-01-13
A day in the life at Kessel Run, one of only 3 DOD programs with an active Continuous Authority To Operate (cATO) https://media.dau.edu/media/Kessel+RunA+a+%22Day+in+the+Life%22+at+a+DoD+Software+Factory/1_y9aexv7y
The others are JIDO (Army) and Platform One (also Air Force)
what is a cATO?
It’s the process for giving an application permission onto a military network
I know this is 8 days later, but we’re hiring for this project in Boston. Actively looking for numerous positions to fill
2020-01-14
@roth.andy maybe still interesting for you: https://www.youtube.com/watch?v=4owbdHzfyMY&feature=youtu.be
2020-01-15
is anyone here actually implementing gitops in their environments? I’m starting to look at it and doubling our repo count is going to be a hard sell.
One other item I’m having generic difficulties with is pipeline as code. Frankly, I’ve got quite a lot of it as I love it. But I keep struggling with developers not allowing me to change it after initial deployment.
I’ve been able to get around this by using a remote git repo of pipeline code as a library of sorts
but I keep thinking I should create a separate protected release/pipeline branch on the repos so I can put separate PR/branch policies in place. Is that some kind of antipattern? It feels like it may be so I’m doubting myself on that one pretty strongly..
We are doing it kinda. Maybe if you can make it to today’s office hours we can talk about it @Zachary Loeber? Otherwise I will prepare a longer answer here.
I’ll be there, just don’t want to dominate the conversation with my issues
Lol, “nerdsniped” - haven’t heard that before
2020-01-16
Just finished a ~40 minute meltdown after updating a prod frontend and noticing the prod site did not work in the same way as the dev/staging versions I had updated no problem.
Turns out the application code is expected to work differently in prod than in dev/staging, with a redirect thrown in on the root domain. And that app code was changed without my knowledge.
There’s no question here, just a quick vent after a very stressful morning where I though I took down our prod homepage.
Yes, that sounds frustrating! Have you guys resolved what to do in the future to avoid this situation?
2020-01-19
after spending some time trying to find a decent documentation platform for a team that doesn’t have confluence and doesn’t want to spend money i’ve been pleasantly surprised by the wiki capabilities in azure devops
markdown and git support, as well as a couple of handy functions such as auto formatting tables and mermaid diagram support
2020-01-21
@Chris Fowles Honestly, the platform isn’t so bad on that front. I like that the wiki is just a repo as well (git push origin wikiMaster)
didn’t realize mermaid support was there, is that recent?
yeh i think so, i hadn’t seen it before
I’d been using Typora locally for mermaid diagram editing live (though VSCode has a live previewer extension that rocks as well)
How badly am I getting declarative-itus if I no longer wish to make visio diagrams by hand anymore?
i don’t mind making diagrams by hand in visio - i hate maintaining them though
2020-01-23
@Erik Osterman (Cloud Posse) how have the Timeular cubes that you got been working? I’ve been using the app for free for a while and like it a lot. If you haven’t been having any issues with the cubes like flaky connectivity then I’m thinking I’ll get one.
I started setting it up but got derailed.
The problem for me is we have a dozen customers with a dozen projects with a dozen tasks
the product wasn’t designed for this, so the UI breaks down.
It has a simple little selector where I have scroll through 12x12x12 options that are all truncated with ...
I don’t think you would have this problem
and aside from that it’s pretty nice
it has somewhat of an “apple” feel to the product and packaging
but some rough edges: the charge the thing is hard. you have to rip off this rubber seal and I damn near pull off my nail in doing so.
2020-01-26
2020-01-27
I’d be super hard pressed to answer all the questions on this list of devops interview preparation exercises… https://github.com/bregman-arie/devops-exercises/blob/master/README.md
Heck, I may start working my way through this list anyway. Its an excellent braindump of technologies and patterns worth reviewing.
These are great questions
Super helpful, thanks. Great resource to look at when I do step in the job market again
2020-01-28
Hello guys, i need help with gcp app engine, i don’t find a good channel for that, is there one ?
Hrmmmm yea, @albttx we don’t have much chatter for GCP topics (mostly an AWS based team)
I’ll create a #gcp channel and we can see where it goes
2020-01-30
Say I converted a typescript from a synchronous (s: string): string to an async (s: string): Promise<string>, and that that function is used hundreds of times in our codebase.
Does anyone have any refactoring tool suggestions I could use to quickly make this change in an automated way?
grep && sed ?
The big issue is chaining.
Sure, it’s easy to change foo to await foo, but then you also need to update the function that used foo to be async if it wasn’t, and have it return a promise, and then find all references to that function, and make them use await, etc. etc. etc.
Sounds like something for a cleverer person than I to code up an abstract syntax tree find/replace tool for….
2020-01-31
Good morning! Quick question (hopefully) I found you searching for a way to mask terraform sensitive output and found tmask, which led me to geodesic. I was inspired to use that as an example to create my own base docker image for a GCP based infra environment. What are your recommendations to include tfmask and github-commenter from your repos? How often are they updated in your flow?
We have not yet had a chance to update tfmask for terraform 0.12
We use the github-commenter everyday. It’s very simple which is why not much activity on the repo.
We use the commenter to add a comment to a PR every time it is deployed to an environment
We comment on the commit
I think the main reason we keep having this conversation is that everyone seems to be arguing from the point of view that it’s either monolith OR microservices. It doesn’t need to be this black and white. You can very successfully have both patterns within the same ecosystem.
Kelsey Hightower rocking the boat again… :-)
“We’re gonna break it up and somehow find the engineering discipline we never had in the first place” < too true.
Not sure about the “bad infrastructure” though (what referring to exactly)
containerizing a bunch of crap but still being so tightly interwoven that they could never be deployed individually.
Just throwing this in here too: https://twitter.com/sarahmei/status/1090860150712221698?s=21
Improvement over consistency is a great mantra