Managing multiple systems in parallel with SaltStack
Pulling the Strings
© Lead Image © perseomedusa, 123RF.com
Professionals often turn to SaltStack to manage server farms in parallel. When used properly, the same technology also saves work in small networks.
Even in smaller IT environments, managing systems as consistently as possible pays dividends. The effort required to learn and use massive tools for configuration management (e.g., Ansible, Chef, Puppet) rarely pays off, because you frequently need to complete simple, one-time tasks; when these present themselves, writing recipes (Chef) or manifests (Puppet) wastes too much time.
Alternatively, many admins use self-made scripts, which they execute to hosts in rotation from a list that is laborious to maintain. Although these methods work, you must take the particulars of the respective platforms into account and cope with possible timeouts while opening connections. This effort can be reduced by turning to a proven solution. The right tool should be able not only to abstract the differences of distributions and operating systems but also to provide its own communication channel and offer modules with pre-built commands and macros. Normally, projects from the orchestration and remote execution environment offer something similar. Thanks to its simple operation, speed, and scalability, SaltStack [1] stands out from the competition in these cases.
SaltStack
The systems and software architect Thomas S. Hatch faced the challenge of centrally maintaining an infrastructure that was partly inconsistent. To master it, he wrote his own software in Python; SaltStack (Salt, for short) emerged and was eventually published as an open source project in March 2011.
[...]
Buy this article as PDF
(incl. VAT)
Buy Linux Magazine
US / Canada
UK / Australia