Happiness can be found, even in the darkest of times, if one only remembers to turn on the light.
In this tutorial you’ll see how to take advantage of the ELK stack to set up a centralized logging mechanism for your Swarm cluster
Funny experiment to make a unified package manager user interface
What it feels like to become a Debian maintainer in 2020
How to gain root access by exploiting wrongly designed setuid executables.
How to gain root access by using a Docker engine running with default configuration.
I used to manage a dozen VPS since many years: Zabbix, Gitlab/Gitlab CI, private docker registry, production environment (3 nodes docker swarm cluster), database server (MariaDB & MongoDB), blog server (running Ghost), logs collector (Graylog), etc… I was spending a consequent amount of money & time for all these VPS, and it was time to change. From Ghost to Hugo One of the most important thing I run is this blog.
I wanted to write a quick off-topic article about the ongoing coronavirus pandemia since there’s a lof of false opinions and fake news on this topic. This virus (SARS-CoV-2) is deadly and not really comparable to the flu, it is at least much more contagious. (We don’t know anything about the death rate yet, since the pandemia is ongoing). You shouldn’t treat it lightly. Not staying home will cause a lot of deaths in the months to come because of the overwhelming of your country healthcare system.
We’ve been experiencing very strange memory leak in our Java application servers at work: when deploying new version of a micro service, the JVM process was running out of memory and consequently crash leading to service outage. After a bit of research, it looked like that these kind of errors were very common with this application server, especially when deploying application without restarting the server from time to time. The common fix was to restart the JVM process before putting in production, preventing any out of memory (but not the memory leak).