SaltStack: Installing an older Salt Master or Minion for compatibility

If your Salt Minion version is too far removed from the Salt Master version, you may find yourself with unexplained errors.

This problem can be faced when the OS template you are deploying was packaged years earlier with an older Salt minion while the Salt Master has been kept up to date.

But it can also happen with a relatively recent version Master like 2016.11, if you use the latest 2017.7 Minion which has major changes in the fileclient.

In this article I will show you how to use apt-get to install an earlier version of the Salt Master or Salt Minion.

Continue reading “SaltStack: Installing an older Salt Master or Minion for compatibility”

GoLang: Vendor directory for github branches other than master

Using 3rd party packages from github is made very simple in the Go language with the import statement.  But one problem is that “go get” will always pull the HEAD of the master branch and there is no way to explicitly specify another branch.

The ultimate answer would be to use a package dependency manager like Glide, which I describe in this article.  But if you cannot introduce Glide into your workflow yet then manually populating the vendor directory (enabled by default since 1.6) is a viable alternative.

Continue reading “GoLang: Vendor directory for github branches other than master”

SaltStack: Installing a Salt Master on Ubuntu 14.04

saltstack_logo-thumbnailConfiguration Management tools like SaltStack are invaluable for managing infrastructure at scale.  Even in the growing world of containerization where immutable image deployment is the norm, those images need to be built in a repeatable and auditable fashion.

This article will detail installation of the SaltStack master on Ubuntu 14.04, with validation using a single Minion.  Note that Minion installation is not mandatory if using Salt SSH.

Continue reading “SaltStack: Installing a Salt Master on Ubuntu 14.04”

ELK: Scaling an ElasticSearch Cluster

elasticsearch-logoThe heart of the ELK stack is Elasticsearch.  In order to provide high availability and scalability, it needs to be deployed as a cluster with master and data nodes.  The Elasticsearch cluster is responsible for both indexing incoming data as well as searches against that indexed data.

Resources

As described in the documentation, if there is one absolutely critical resource it is memory.  Keeping the heap size less than 32G will allow you to use compressed object pointers which is preferred.  Swapping memory takes a big hit, so minimize swappiness on your Linux host.

Continue reading “ELK: Scaling an ElasticSearch Cluster”