GCP loadbalancer monitoring, aggregated per route

Hey guys,

Would you know of a tool, SAAS preferably, which would read logs from a GCP loadbalancer and would produce stats (latency/volume/error) aggregated per routes?

By route aggregation I mean:

GET /api/users/52356

GET /api/users/1234

I'd love the tool to be able to detect those two routes are actually the same: /api/users/{\\d+}

I cannot find anything like that, so I made something like LB logs -> big query -> custom view with route parsing -> google data studio to visualize.

If it doesn't exist, should we build that? :)

https://redd.it/llyw0f
@r_devops

So, here's a question...

Is every DevOps engineer in Romania taken?

Honestly, I know the War for Talent is real, but it seems as though for each member of the DevOps community, there are at least half a dozen job offers lying around.

Anyway, there are a few projects (quite a lot, actually) we are working on and we need like lots of great DevOps engineers and maybe you could help me with a few pointers on what is truly attractive to you, when considering job opportunities. Any sort of information is priceless, right now, and greatly appreciated!

https://redd.it/llxbae
@r_devops

Is kubernetes-external-secrets mature enough?

We seek a solution to fetch secrets from various kms/secrets-manager (e.g. aws secret manager) into our k8s cluster as secrets. kubernetes-external-secrets seems to satisfy our requirements but is it mature and stable enough based on your experience?


https://github.com/external-secrets/kubernetes-external-secrets

https://redd.it/llxmmp
@r_devops

CICD pipeline - Gitlab and gke...and helm?

hi - looking to start this, so wondering how everyone deploys to gke using gitlab? i dont have any experience with either.

i am thinking the following - most people build and test in a gitlab pipeline => if tests pass push to a repository with a tag of test (or similar), and then use helm to deploy to a k8s staging environment. once the container is in k8s staging, run functional tests...but then how do you get it in to prod? if functional tests pass, tag with prod and then do a helm deploy again?

any insights in to what people are doing/have found useful/wouldnt do or would change if doing it over again would be awesome!

cheers guys!

https://redd.it/llr2yr
@r_devops

Manifest tag update w/ GitOps workflow

I'm curious to see how others are handling the process of updating manifest files when new tags are pushed for images when following GitOps practices.


You've got your various application repositories - they are responsible for building -> testing -> pushing the image and tag up to the image repository. Then you've got your manifest repository that holds your Helm charts or Kustomize or vanilla yaml, etc., that something like ArgoCD or Flux is watching for changes to update in the cluster. What is the preferred approach/best practice for linking the tags created by the application repositories to the manifest repo?

https://redd.it/llu7dy
@r_devops

Hackerrank for a devops role

I just finished a hackerrank "test" for a devops role. It was a pretest before the interview. I've never used hackerrank, I've always viewed it more for programming. Coming from a sys admin background, yeah I can code a bit, script stuff absolutely, build a pipelines for sure. Make pictures about "devops" architecture and answer obscure questions on msg brokers, and hardly used git commands. There was a bunch of crap stuffed in there that left me scratching my head, not that I didn't know it or have a partial answer. It was just by the book multiple choice answers to pick from, you know the kind that makes it look like a trick question. Not to mention you don't get to use Google or anything you script you can't write print statements to debug, it's basically right or wrong. I love too they don't give you any background to what the topics are so you just go in blind.

I appreciate questioning skills, giving a homework assignment. But I just don't feel like hackerrank was a good option for a devops role.

Anyone have a similar test or experience with hackerrank for devops?

https://redd.it/llmzue
@r_devops

CI/CD pipeline for database changes

Hello World !

^(First time posting on Reddit ! 🆕)

Here an article to include database changes in your continuous delivery process !

https://medium.com/tales-of-libeo/continous-integration-make-sure-database-changes-are-included-using-gitlab-ci-cd-6191e984f8d0


It's done using GitLab CI and PostgreSQL, but no matter what you use, concept should be kinda the same ! Looking for some feedbacks ! 🤗

https://redd.it/llqyyg
@r_devops

What size server for two mobile and two web apps?

My company is planning to launch two mobile and two web applications, all of which have dynamic, not static content. Collectively they make up an on-demand platform, and we’re wondering how big a server we should purchase in terms of memory transfer, and SSD space.

We plan to have one server for each application, and don’t expect to have too much traffic initially since we’re launching in one city. Support for ~1,000 users would be ideal.

https://redd.it/lm9dnb
@r_devops

Terraform and Jenkins

Hey Guys,

I just wanted to ask you for an advice. If you have project in Terraform which is broken into multiple objects(per resource set) - for example:

\- Main Core Virtual Network, NSGs, Subnets, FW etc. all configured in one configuration file

\- Resource Groups - all configured in separate configuration etc.

Same will apply for other resources. Each configuration has its own idependent state file. My question is: how you would go about Jenkins pipeline configuration. Would you create pipeline per resource or you would somehow use one Pipeline. There is a possibility to convert everything into modules and run everything from one main file configuration file. Would that be a solution? So, if one module would change it would only apply the config based on that changed module. Is my thinking right here?

https://redd.it/lm6jlz
@r_devops

Terraform - Warn for hitting AWS Service Quotas before apply

The Problem: terraform apply creates resources without taking into consideration services quotas.

The Need: Verify there are enough Service Quotas before terraform apply creates new resources.

Background: Staging and Production environments have a dedicated AWS account (one for each). Successful plan execution in Staging does not mean it will succeed in Production since they both hold different Service Quotas limits.

Offered solution: Add to the terraform-aws-provider a piece of code that can warn or fail the terraform apply execution due to hitting a Service Quota's limit. This also means adding additional code to terraform that adds two flags -service-quotas-warn (show warnings) and -service-quotas-fail (fail if there's a warning).

DISCLAIMER: It's possible to enforce the same Service Quotas across multiple AWS accounts with AWS Organizations, though it's irrelevant since not all Terraform users use the AWS Organizations service.

Example:

$ terraform apply -service-quotas-fail -auto-approve
...
[LOG] Checking quota limits
[WRN] VPC - required 6, quota 5
[WRN] EIP - required 17, quota 15
[WRN] Request the above Service Quota before proceeding
[ERR] --service-quotas-fail was set # exit 1

My Question To You Is: Is this something you would use? Have you also experienced issues when deploying to Production and hitting the Service Quotas limits?

https://redd.it/lm5w3o
@r_devops

Experiences with Macbook M1?

Hey,

Is anyone using a M1 Macbook for working in a DevOps/SRE role?

I'd like to read some experiences regarding the most common tools:
- Docker
- Terraform
- Packer
- Ansible
- Kubernetes (kubectl, etc.)

I've done some research but didn't find a DevOps point of view yet.

What really attracts me about these new Macbooks is the battery life paired with the great performance.

Thanks in advance!

https://redd.it/lmc986
@r_devops

Does this sound like devops?

I received an invitation to sit for an interview for a Devops Trainee position but the job description sounds so vague and different from what I googled about devops. I'm from a non-tech background and it was a role I randomly applied for without much consideration because I didn't actually think I would get an interview. Now I'm wondering if it's a clickbait job title to draw in inexperienced applicants and if the interview would be worth my time. My goal is to eventually to transition into tech but I'm not sure if this is the right way forward. Would some appreciate some advice.

​

Work with tech team to develop a deep understanding of the business cases for change, then drive the approval of projects through site and IT governance processes
Manage site IT Risk, Change, & Compliance tasks for daily operations and during project deployment
Maximizes business value via technical knowledge of strategic IT Applications
Key role in promoting good information practice, utilising technology and improving business processes.

https://redd.it/lmdeci
@r_devops

Image Pull Error K3s

Hey guys,

I am trying to debug the below issue while deploying MariaDB 10.1.14 version on the cluster.

pulling image: rpc error: code = NotFound desc = failed to pull and unpack image "docker.io/library/mariadb:10.1.14": failed to unpack image on snapshotter native: failed to extract layer sha256:42755cf4ee95900a105b4e33452e787026ecdefffcc1992f961aa286dc3f7f95: failed to get reader from content store: content digest sha256:5c90d4a2d1a8dfffd05ff2dd659923f0ca2d843b5e45d030e17abbcd06a11b5b: not found

We are using the native snapshotter because of the environment feasibility.

Any leads on debugging are highly appreciated.

Thanks

https://redd.it/lmgjvb
@r_devops

The Reports of Devops's death are greatly exaggerated

Hey everyone. I wrote a post about the transformation of devops:

https://adnaan.badr.in/blog/2021/01/25/the-reports-of-devopss-death-are-greatly-exaggerated/

This covers my experience and doesn't comment on every aspect of devops out there. Would love to hear some thoughts.

https://redd.it/lmhtim
@r_devops

Any one with good spark kafka experience

Spark kafka automation managing tons of different users with high eps any thoughts tips

https://redd.it/lmcsg7
@r_devops

Blog A CLI tool that can easily manage more than 281 runtimes, making it easy to keep things updated and/or switching between versions. And it’s open-source :)

I just wrote this post. I'm showing an OSS tool that can easily keep things updated on *NIX/Windows and/or easily switch between versions. It's very useful on CI/CD pipelines.

Feedbacks/suggestions/* are always welcome :)

See on Medium: https://medium.com/bash-tips-and-tricks/an-easy-way-to-switch-between-runtime-versions-nodejs-terraform-and-279-53fdfbcb4049?sk=43548418bf394d80e277a39a76070843

See on my Website: https://www.lozanomatheus.com/post/an-easy-way-to-switch-between-runtime-versions-nodejs-terraform-and-279

https://redd.it/lm57x5
@r_devops

WireGuard for fleets?

I've been wanting to see if WireGuard might work for our fleet, but I honestly can't see how it can work at a large scale if you need to generate a keypair for every host and drop the public key into the server's config.

I guess a single key could be reused across hosts but that seems insecure.

We currently use OpenVPN with TLS auth, so all we need is a signed cert on the client's side to authenticate which works quite nicely. But WireGuard supports no such thing.

I've also seen paid solutions like Tailscale but I would have a hard time convincing the company to fork out enough money given our fleet size. Also checked out ZeroTier, but despite being open source, their licensing means you have to pay for their network controller. Not affordable for our fleet size.

https://redd.it/lm15u2
@r_devops

Nextjs10 & Serverless Framework

Working on a boiler plate project for serverless framework & NextJs10.

This seems to be the latest plugin: https://github.com/serverless-nextjs/serverless-next.js

I don't seem to be getting anywhere following this simple guide.

Does anyone have any advice or material I could use to get off the ground with this please?

https://redd.it/lml2rp
@r_devops

NodePort issues

Hi there,

I am performing some testing on MySQL pod using sysbench running on a remote server.

I tried to mount the NodePort service to the host but it doesn't work, see below the criteria and results.

k get pods -o wide

NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES

mysql-7d9bcb665c-cdkfx 1/1 Running 0 3h36m 10.42.0.36 k3s-node2 <none> <none>

[root@testsytem\]# k get svc

NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE

mysql NodePort 10.43.84.78<none> 3306:30016/TCP 22m

From the host, if I curl to the pod cluster IP with the port, it works fine but the Nodeport

curl -Iv 10.42.0.36:3306

* Trying 10.42.0.36:3306...

* TCP_NODELAY set

* Connected to 10.42.0.36 (10.42.0.36) port 3306 (#0)

\> HEAD / HTTP/1.1

\> Host: 10.42.0.36:3306

\> User-Agent: curl/7.65.3

\> Accept: */*

\>

Results for nodePort

curl -Iv localhost:30016

* Trying 127.0.0.1:30016...

* TCP_NODELAY set

* connect to 127.0.0.1 port 30016 failed: Connection refused

* Failed to connect to localhost port 30016: Connection refused

* Closing connection 0

curl: (7) Failed to connect to localhost port 30016: Connection refused

I see on the ports on the host, nothing listens on 30016?

How can I debug this issue?

Thanks

https://redd.it/lm4c4n
@r_devops

Hashicorp packer on macbook M1?

I recently decided to invest on a laptop and dedicate it on development and devops stuff..


I thought it would be a good idea to get the new Macbook M1 pro.. It could be a mistake..


I tried installing Packer through brew and it does not find a bottle.. at first I was hoping it was an issue related to Big Sur but then I realised it works just fine in all of my other macbooks that have an Intel cpu.


Has anyone had the same issue and managed to solve it ?


Thanks!

https://redd.it/lm3a5z
@r_devops