Printing

Probably like many amongst us the time of the Christmas holidays is perfect to get some IT related stuff back on track. I used to have a print server setup which got broken over time and I didn't found the energy to invest time into fixing it. But the pressure became higher and higher. 111 From both my wife and daughter, especially during the holidays where the wife want to use it to print out tickets and the daughter want to print out color plates..

So during one of the evenings I pulled myself together and installed ArchLinux ARM …

Since a few years now I'm a happy Archlinux user. I like their philosophy which was one of the major points why I made the switch back in the days.

I'm not only using it on my laptop, but do have some devices running at home which are configured with it. From a thin client which I use as a docker node through some raspberry pies running ArchlinuxARM.

Since Arch is a rolling update distro there are several updates available throughout the day. To keep on top of them I had to log in on all those devices at least …

bumping into the case where once deployed a full stack application we don't have any direct connection due to no uplink for security reasons.

So we (you too @Tom) looked into a way to export the prometheus data into a tar.gz which could be transferred and imported into an instance on our local machine.

After the initial blog post where we created a tar.gz file from the prometheus storage.tsdb.path on the filesystem @roidelapluie pointed me out about the snapshot feature.

So we did a bit of research and came up with this new procedure.

First of …

as published a few months ago I worked out a dockerized a jenkins farm where both master as slaves are docker containers working together with services like nexus and such. Next to that setup I've dockerized my home setup where services like pi-hole, home-assistant and others are running as docker containers on a thin client I promoted to my home lab.

To have an overview about all those containers and the resources they are consuming I pulled in the git repo of Brian Christner which spins up a whole prometheus stack with some exporters and a grafana instance to visualize …

a few months ago I configured a thin client as my home server to replace the previous raspberry pi setup.

During that migration I moved over all native services within docker containers. One of those services being a pi-hole setup to block ad serving domains on dns level and to have a dns cache within our LAN to gain a bit of speed.

It has been running ever since without any issue and worked pretty well.

When cloudflare announced their fast and privacy based DNS resolver I got a bit intrigued by their DNS over HTTPS feature. Especially since our …

When refactoring a check_memory I wrote a few years ago I bumped into the feared

NRPE: Unable to read output

error message on our nagios instance.

When looking for a solution I went through most possible debug steps I could think of and which are nicely described by nagios support but didn't found any solution.

I almost grabbed to some anti depressants when I thought of the thing I always forget about.

SELINUX

When crawling through the audit log it became clear I forgot to configure the proper selinux context type for the new script.

type=PATH msg=audit(1518702310 …

recently I started working at a new project where the infra is maintained by ansible. When been asked to write some functionality in a playbook I missed my vagrant puppet setup where I could easily test my puppet code on my local machine.

Due to my previous project I felt like maybe I could use docker for this purpose on the ansible part. So I looked a bit around and stumbled on the docker-ansible github repository of William Yeh. He already did a great job by creating a docker container with ansible preinstalled for a lot of linux distributions.

I …

in a previous blog post I talked about setting up a dockerized jenkins master/slave setup and setting up a private docker registry using nexus.

The next thing on the roadmap was to use this jenkins setup to actually build new docker images for specific software. Before going to the different teams and talking how they now build their software and how this could be done using this new containerized setup I setted up a new jenkins job.

This jenkins job will build a generic jenkins slave docker container which will be used by the jenkins master to build some …

Today we bumped into an interesting issue in the jenkins builds of some android based applications. The gradle commands succeeded but then suddenly failed the build with this most cryptic message ever:

BUILD SUCCESSFUL

Total time: 1 mins 20.492 secs

FAILURE: Build failed with an exception.

* What went wrong:
Already finished

* Try:
Run with --stacktrace option to get the stack trace. Run with --info or --debug option to get more log output.
[Pipeline] }

Since this came out of nowhere without any modification on the build servers we where flabbergasted since the builds ran fine on our local machines.

So …

looking for a global repository store which could store maven projects, yum repositories, docker repositories, we bumped into Nexus repository manager. We used the official docker image to see how it can be implemented in the dockerized CI environment.

docker repository

as a first the docker repository feature could be enabled so we can start building and storing docker images for the different jenkins build slaves and the jenkins master so our work is reproducible and stored in a safe central place.

We configured 3 repositories in nexus for our docker images seen as a recommended approach in the nexus …