azure app services - containers deployment

Hello everyone,

recently I've got an issue with one func app and one web app, both linux. the old deployments was packing the app as a zip and deployed on those 2 app services. my issue came after I tried to deploy as a container. on deployment history, and on portal it's clearly says that was deployed from container. even the app service dont startup with the wrong docker credentials. but i have found that those app services are still reading from the old .zip that remained on those app services even of i deploy as a container.

does anybody encountered this from switching the deployment mode from . zip to container? did you find any solution?

https://redd.it/1m7vt8o
@r_devops

Suggestions and review

I am trying to get into devops role, currently i am working in WITCH in my current role i am working on automation framework which is in python. I have not completely real world experience for devops but in my current project is use of github actions and jenkins so i have been learning these two alongwith docker and kubernetes. For past 3 months. I have prepared a resume but my resume is not even getting shortlisted to at least give test or interview. Please suggest if there is anything that i should update to my resume.


https://www.dropbox.com/scl/fi/cczcuu47rlognrose3cit/IMG20250724114919.jpg?rlkey=nw1c97dlfn7fcerplqybz8h2l&st=nkhiwm8b&dl=0


https://redd.it/1m7wv4d
@r_devops

How we automated Unity editor workflows with AI — open devops story behind the scenes

Hey folks 👋

I wanted to share something from the devops side of a project I’ve been building called Code Maestro. It’s a Unity-based tool designed to speed up repetitive editor tasks — like auto-fixing scripts, organizing assets, placing prefabs, and more. While it’s technically a Unity tool, it also became an interesting devops challenge in terms of automation, tooling, and team-less CI-like behavior.

🔧 DevOps angle:

No external server — everything runs locally as a standalone AI-powered assistant
Uses a custom lightweight task queue internally — almost like a mini CI/CD pipeline for Unity scenes
We explored different ways of triggering agents from scripts, GUI buttons, and even shell calls
Built internal logging/feedback layers to self-assess output correctness (kinda like unit tests for AI decisions)

💬 Would love to hear your thoughts:

If you’ve ever done internal tooling in Unity or another engine — what was your approach to testing / automation?
Any favorite practices for dealing with AI or agent-based flows in game dev or scripting pipelines?

The code and experiments are open:
🔗 GitHub: https://github.com/codemaestroai/advanced-unity-mcp

P.S. I’m not promoting anything commercial — just looking to exchange ideas with people working on internal tools, AI workflows, or Unity devops

https://redd.it/1m7xpbw
@r_devops

Process vs autonomy/trust

I read this article from an engineer who worked as an SRE at Google for 16 years and this stuck with me:

>More process doesn’t mean more control, it usually just means more friction

It was surprising, I imagined a massive company like Google would be full of processes to keep things safe and would promote processes.

Setting up processes makes me feel at ease tbh. Most of the time it works. But as things get more messy, keeping track of the many playbooks etc is difficult. I feel it keeps getting harder for me to even know if they're still relevant. But where do you draw the trust line ? How rigid should safeguard rails be?

An 'it depends' question of course but I'd like to hear your thought process on this

ps. the article is more centred on this thinking process for incident management but if you want to check it out it's this one: https://rootly.com/blog/when-process-becomes-latency-optimizing-incident-response-cadence

https://redd.it/1m7yuuu
@r_devops

I'm a full stack software engineer who want to transition to devOps.

I have 1.5 YOE as a software developer as of now based in India. In my current role im using a lot of aws microservices and learning CI/CD,IaC and all. with my experience level is this possible to get a job in devOps field?? also wherever i get the video tutorials and they all seem like you literally need each and everything from that tech stack to really get a job,is this true? I need guidance on how I should proceed with all this.

https://redd.it/1m82zvq
@r_devops

AI FOMO - is anyone using AI at work beside writing code?

I use Claude for kick starting a lot of my projects and scripts, but is there another way of using AI to my advantage? Some things that specifically come to mind:

- n8n is popping everywhere. Did anyone automate some workflow with it in a meaningful way?
- Logging and error analysis?
- IaC reviews?
- CI/CD optimizations

I want to specifically focus on the "bring your own AI" part, instead of relying on new SaaS stuff to buy or implement.

Any ideas or fun projects would be nice to learn from.

Thanks!

https://redd.it/1m8461p
@r_devops

Web Dev

hello guys , hope you are all good

i want to ask about web dev cause i heard that i will need to learn front end from somme people for the 2nd year CS , so what i should learn and is it really that i will not need html , because i started to learn it

at the end , thank you to every one that responded to me

https://redd.it/1m88sx6
@r_devops

Technical interview with food delivery company

So I passed the initial screening interview and now have the first technical interview scheduled for a company I can’t name yet that has a known food delivery app. I have around 5 years of DevOps experience, and a good knowledge of most of the tools of the trade (docker, kubernetes, terraform, ansible, helm, kustomize, argocd…). Thing is, I never worked with mobile apps so I’m looking for any advice on what to prepare outside my scope or on how it can be different for me.

https://redd.it/1m89fz2
@r_devops

End to end CI/CD pipeline for a C application

I know the interwebs are chock a block with pipelines for Java/python, but I am an programmers who still loves his C. Recently after being away for several years due to personal reasons, I have taken up a C project for a client. Just wanted to know about the opensource options for an end to end CI/CD pipeline for a C project.

Github > Jenkins > GCC > sonarcube > trivy > Cmake or Ninja > Nexus > docker > kubernates

Is this correct ? My doubt is whether GCC and CMake can be integrated as part of this pipeline. Reason is for Java there is Maven. Do we have something for C that compiles and builds similar to maven?

Any help is most appreciated. Much obliged.



https://redd.it/1m8be6y
@r_devops

How to Sort Work Item View Differently

Hi All,


Does anyone know how to change the sort criteria to match the order of Work Items within the Backlog tab? We have resources that are utilized across multiple backlogs (I've told management many times this is a bad idea), so as a work-around, we've updated the Priority Field for our Features (and related Stories, Tasks, and Bugs) to hold the same priority, this way when someone comes into the Work Item view, they can see all of their work (via the Assigned to Me dropdown), prioritized from highest to lowest priority. The problem is, the Work Item view keeps moving around items within the same priority by which one has been most recently updated, which is not necessarily their top item to work on within the Feature. Any help is greatly appreciated, thanks! <3

https://redd.it/1m8e1rf
@r_devops

I turn it into a server from my laptop at my own home.

I'm making a software for it. I welcome your suggestions.
Pls Click -> Blog <- Pls Click

https://redd.it/1m8gpqc
@r_devops

DevOps professionals - I need your insights!


Hi everyone ☺️ I'm a postgraduate student researching racing to prove why DevOps adoption in large organisations (such as AWS, Microsoft, Google, Meta, etc) sometimes fails to match the hype.
I call it the DevOps Implementation Paradox (DIP) framework: companies adopt DevOps for prestige or branding, but face real struggles with legacy systems, culture and leadership misalignment.
For research, I'm running a quick survey (anonymous) to capture real-world challenges and enablers from engineers, SREs, DevOps leads and anyone working within this field or with CI/CD pipelines.
Your input will help expose the gap between DevOps hype and practical reality 👏🏻 and will be used ethically in my dissertation.

Copy survey link here:
https://docs.google.com/forms/d/e/1FAIpQLSf17Bd_kAM7G7OTeGIdq5Vcy-uGWlJ3NNaj1qzqFLKBzxkvjw/viewform?usp=header

Thank you for helping bridge the DevOps reality gap! Happy to share final insights with anyone interested.

https://redd.it/1m8h1rx
@r_devops

Using AI as a security coach in workflows

Yes, AI bad. Don't rely on it. It hallucinates. I agree with all of that. But please hear me out.

We're an ultra tiny shop. And our dev team is junior heavy. It's not an ideal situation. They consider things to be done if they work and don't always consider security implications. On review, we found a pretty glaring privilege escalation vulnerability in one of our APIs.

We're already running Snyk scans on code, but stuff like this slips by. And yes I know human review and other tools are fairly effective, but time is short and people miss things.

So, today I hopped into AI foundry and wrote a prompt and ran some sample code through it that I know is problematic. The initial results are promising and I intend to attach it to workflows for running against our critical micro service APIs when they change.

Before I do that, I wanted to get some feedback. I am working from the angle that I want it to scan subsets of the code and make sure good practices are being followed (authentication, tokens, etc) but I don't want to write the code for the dev. Because hallucination. For web apps, bounce it against things like OWASP top 10 rules, tell you where you screwed up, give a leading suggestion, but don't give a "here's the full fix" snippet. Because I want the devs to actually learn. And I want humans to remain firmly in the loop.

Does this sound like a good approach? If you've done this before, can you share any gotchas?

https://redd.it/1m8j7z8
@r_devops

Is the KubeCon worth attending?

I am a senior Devops. Not sure what I can get from KubeCon. Also interested in ArgoCon this November.

https://redd.it/1m8p5ec
@r_devops

Cloudflare wildcard certificates

Hi everyone,
I recently switched to using Cloudflare certificates (with DNS proxying enabled) and a wildcard cert for my domains. Just wanted to ask:

* Is this generally considered good practice?
* What are the pros and cons of using a wildcard cert with Cloudflare?
* Are there any security or scalability concerns I should be aware of compared to using individual certs?

Thanks in advance!

https://redd.it/1m8qsun
@r_devops

Why MCP(Model Context Protocol) Matters for Your AI Projects


I wrote my first article on MCP


https://www.linkedin.com/pulse/why-mcpmodel-context-protocol-matters-your-ai-projects-deepak-kumar--fmbfc?utm_source=share&utm_medium=member_android&utm_campaign=share_via

https://redd.it/1m8snzd
@r_devops

Anyone using XDR for cloud-native threat detection?

We’ve shifted most workloads to ECS and Lambda, and our old endpoint tools don’t cover squat anymore. I keep hearing about XDR as the next-gen detection approach, but it feels like half the vendors define it differently.

What are you using to detect lateral movement, container escapes, and other cloud-native threats?



https://redd.it/1m8tecu
@r_devops

How to upskill?

I currently have Azure fundamentals cert and CKA. Wondering how to upskill next? Is redhat administrator cert worth doing?

https://redd.it/1m8sguk
@r_devops

Trusting the Boot Process: Inside Bottlerocket's Security Architecture

https://molnett.com/blog/25-06-30-trusting-the-boot-process

Bottlerocket is a distro developed by AWS for their more sensitive container-based environments like AWS Govcloud, EKS anywhere and others. We thought it would be a good choice for us (we're building a EU-focused Serverless cloud) as many of our customers are in Healthtech, so we've used it for all our nodes, even the Kubernetes control plane.

My colleague Mikael decided to dive deeper into how the boot process works, and in a later post how it interacts with the TPM.

I would love to hear how (and if) you've solved this for your own platforms, and if so what you think of it!

https://redd.it/1m8vewk
@r_devops

DDoS attack - i think

I manage several ecommerce websites and their hosting for work. Over the years I have seen various types of attacks, as well as an increase an AI / bot traffic.

On the 3rd July I was alerted to high server activity on one of our sites. When I was reviewing the server and nginx logs, I could see the requests per hour to the site had gone from an average of 20,000 an hour to 120,000. However Sales had not increased,

Reviewing the nginx logs, I found that there was a large number of requests to a small group of category pages, never any request for CSS / JS - which stinks of bot.

Cherry picking some IP addresses, they only ever made one request.

Immediately we enabled cloudflare under attack mode, which made the traffic instantly drop, adding to the idea that this is bot traffic and not a successful marketing campaign.


I identified patterns in paths and created a rule in cloudflare to target this, allowing me to remove the under attack mode and keep the website online.

Between then and now I have been reviewing the requests hitting my rule.

A few times I downloaded and analysed 500 requests to the rule and they all read similar to this.

\- 493 Different IP addresses
\- 278 ASNs
\- 55 Countries
\- 13 URLs
\- 412 User Agents
\- 500 different query parameters


The website sells items to the UK, a large number of these requests are coming from Brazil, Singapore, Vietnam, India and Bangladesh


Checking on the rule today (25th july) so 3 weeks in - and within cloudflare I can see the rule is blocking a LOT of requests. This is showing is has presented the challenge 18k requests in the last 24 hours.

I should add, my rule is set to ignore for known bots.


Is this a DDoS Attack? I have never had one this sophisticated or last this long.

The website is not high value and the requests have been blocked for 3 weeks now yet they still continue to come in.


Any suggestions on additional things I can do to tackle this would also be welcome



https://redd.it/1m8vf0n
@r_devops

Please help me with nifi and nifikop that i'm trying to learn!

I encounter a few problems. I'm trying to install a simple HTTP nifi in my Azure Kubernetes.
I have a very simple setup, just for test. A single VM from which I can get into my AKS with k9s or kubectl commands.
I have a simple cluster made like:

```
az aks create --resource-group rg1 --name aks1 --node-count 3 --enable-cluster-autoscaler --min-count 3 --max-count 5 --network-plugin azure --vnet-subnet-id '/subscriptions/c3a46a89-745e-413b-9aaf-c6387f0c7760/resourceGroups/rg1/providers/Microsoft.Network/virtualNetworks/vnet1/subnets/vnet1-subnet1' --enable-private-cluster --zones 1 2 3
```

I did tried to install different things on it for tests and they are working so I don't think there may be a problem with the cluster itself.

Steps I did for my NIFI:

1.I installed cert manager,
```
kubectl apply -f https://github.com/jetstack/cert-manager/releases/latest/download/cert-manager.yaml
```

2.
zookeper,
```
helm upgrade --install zookeeper-cluster bitnami/zookeeper \
--namespace nifi \
--set resources.requests.memory=256Mi \
--set resources.requests.cpu=250m \
--set resources.limits.memory=256Mi \
--set resources.limits.cpu=250m \
--set networkPolicy.enabled=true \
--set persistence.storageClass=default \
--set replicaCount=3 \
--version "13.8.4"
```
3.
Added nifikop with servieaccount and a clusterrolebinding,
```
kubectl create serviceaccount nifi -n nifi

kubectl create clusterrolebinding nifi-admin --clusterrole=cluster-admin --serviceaccount=nifi:nifi
```
4.
```
helm install nifikop \
oci://ghcr.io/konpyutaika/helm-charts/nifikop \
--namespace=nifi \
--version 1.14.1 \
--set metrics.enabled=true \
--set image.pullPolicy=IfNotPresent \
--set logLevel=INFO \
--set serviceAccount.create=false \
--set serviceAccount.name=nifi \
--set namespaces="{nifi}" \
--set resources.requests.memory=256Mi \
--set resources.requests.cpu=250m \
--set resources.limits.memory=256Mi \
--set resources.limits.cpu=250m
```

5. nifi-cluster.yaml
```
apiVersion: nifi.konpyutaika.com/v1
kind: NifiCluster
metadata:
name: simplenifi
namespace: nifi
spec:
service:
headlessEnabled: true
labels:
cluster-name: simplenifi
zkAddress: "zookeeper-cluster-headless.nifi.svc.cluster.local:2181"
zkPath: /simplenifi
clusterImage: "apache/nifi:2.4.0"
initContainers:
- name: init-nifi-utils
image: esolcontainerregistry1.azurecr.io/nifi/nifi-resources:9
imagePullPolicy: Always
command: ["sh", "-c"]
securityContext:
runAsUser: 0
args:
- |
rm -rf /opt/nifi/extensions/* && \
cp -vr /external-resources-files/jars/* /opt/nifi/extensions/
volumeMounts:
- name: nifi-external-resources
mountPath: /opt/nifi/extensions
oneNifiNodePerNode: true
readOnlyConfig:
nifiProperties:
overrideConfigs: |
nifi.sensitive.props.key=thisIsABadSensitiveKeyPassword
nifi.cluster.protocol.is.secure=false

# Disable HTTPS
nifi.web.https.host=
nifi.web.https.port=

# Enable HTTP
nifi.web.http.host=0.0.0.0
nifi.web.http.port=8080

nifi.remote.input.http.enabled=true
nifi.remote.input.secure=false

nifi.security.needClientAuth=false
nifi.security.allow.anonymous.authentication=false
nifi.security.user.authorizer: "single-user-authorizer"
managedAdminUsers:
- name: myadmin
identity: [email protected]
pod:
labels:
cluster-name: simplenifi
readinessProbe:
exec:
command:
- bash
- -c
- curl -f https://localhost:8080/nifi-api
initialDelaySeconds: 20
periodSeconds: 10
timeoutSeconds: 5
failureThreshold: 6
nodeConfigGroups:
default_group:
imagePullPolicy: IfNotPresent
isNode: true
serviceAccountName: default
storageConfigs:
- mountPath: "/opt/nifi/nifi-current/logs"
name: logs