Passing client cert through nginx to the backend

Recently, I’ve had to work on a certificate authentication in our system, where we use nginx as the proxy to all of our services. Our services talk to each other in mTLS connection(https://en.wikipedia.org/wiki/Mutual_authentication), and the challenge here were 1. How do we go about passing in the client cert through the nginx to the backend? […]

My new toy: Raspberry pi 3 Model b+

I’ve recently bought a raspberry pi 3 model b+ which became so popular once they’ve released it as a celebration for the pi-day(3/14) this year. It’s my first personal rasPi, though I’ve always been wanting to get one for my self even after college. Wanted to share some tips how to set up the rasPi […]

Monitoring a containerized app with JProfiler

I had to figure out what impact my change would have on the memory usage of the application. With my workspace setup, where I have a main method I could call the class in Eclipse, I’ve installed Jprofiler(https://www.ej-technologies.com/download/jprofiler/files) on my machine, which profiles the java metrics, which is good for cpu or memory analysis. I’ve […]

Size inducing by measuring global usage of a cluster

Measuring the global usage of distributed application is necessary for estimating horizontally scalable infrastructure needs. Kubernetes provides a mechanism for multi-tenancy by its namespace feature. A quick and dirty way to estimate cluster bin packing can be accomplished by simple for loop like this. END=100 for i in $(seq 1 $END); do SECONDS=0 name=”claire-$i” kubectl […]

Config map creation could happen in one yaml

We could just add a configmap key value pairs into one yaml file along with the other orchestration (service definitions). Adding a following block on top of the yaml file that we used for the orchestration would create a configmap named hpup-admin’ under whichever namespace defined by a user. For example, when having to mount […]

Gaining insights from Prometheus with creating a service under a different namespace

My last blog post talked about configuring the nodes to export the system level metrics to a master prometheus service. With the setup we could now determine the cost of creating another cluster (with a different namespace) There are three different things that I find essential to Kubernetes. 1. Namespace 2. Configmap 3. Yaml file […]

Setting up node_exporter for nodes for extensive metrics measurement from Prometheus on kubernetes cluster

Prometheus is an open source library for collecting metrics from any system. Especially in distributed system, it could get difficult to measure things in one, but thanks to Prometheus, we could grab metrics from each node and gain some insights on scalability and etc. Node exporter is used to export the system level metrics on […]

Back to using pandas and Jupyter

I’ve been missing using pandas and jupyter for data analysis and machine learning. For the hackathon at my company, I chose the machine learning/data engineering project to re-polish my knowledge on those tools. Have set up my Jupyter 3 with anaconda, and got started by testing the prometheus end points. Prometheus(https://prometheus.io/) is a metric collection […]

Cfssl using authsign endpoint.

I’ve been trying to use the authsign endpoint on the cfssl (cloudflare’s PKI TLS toolkit)  However due to the lack of documentation for the specific endpoint it was not really intuitive to use it. I had to reverse engineer a bit and used Go to prototype.   Once you have the json constructed, you can send […]