Installing Niginx is very easy and straight forward with yum on CentOS. But I found it stopped to work last night. My installation script aborted with an error saying that it could not find libunwind package, which turns out no longer in the EPEL7 after a upgrade a couple of days ago.
I started to play with CentOS 7 recently because I wanted to use it for one of my appliance products. Naturally, I installed the minimum version of the CentOS 7.0.
While it takes time, the installation process is pretty straight-forward like installing any other Linux on a virtual machine. What I did was to upload the .iso file to a datastore, then create a new virtual machine that uses the iso file as CD device. When it booted, the installation started. Although it’s minimal version, but the installer is actually GUI based. So it’s pretty intuitive.
Like RDP, VNC protocol allows for accessing computers remotely with full window GUI. It’s used in VMware vSphere and OpenStack for users to interact with virtual machines by connecting to hypervisors. Here is what I just did to set it up. The instruction should work on a host either it’s a physical machine or virtual machine as long as it’s installed with CentOS or equivalents.
Installing VNC Server
While using PackStack to install OpenStack for multi-node topology, I found my SSH client was so slow that it failed the PackStack installation command. The ssh I had was the default one with CentOS 6.4: “OpenSSH_5.3p1, OpenSSL 1.0.0-fips 29 Mar 2010.” It seemed to work just fine while using PackStack for all-in-one deployment as described in my previous article.
To isolate the problem, I started to print more debug information from the SSH as follows:
Having successfully installed OpenStack all-in-one with PackStack, I started to try out the multi-node deployment. It ended up much longer time than I thought because of various issues mainly with networking. The following summarizes what I did to make it work, and some tricks and tips I found out during the process.
With increasing popularity of OpenStack, I finally got chance to try it out by myself. As the first step, I chose to install everything into one machine (update: for multi-node deployment, check here). Thanks to the packstack project and RedHat documentation (http://openstack.redhat.com/Quickstart), it’s supposed to be straight forward and painless on CentOS. It turned out to be true if and only if you install it with direct HTTP(s) access to the Internet.
In my last post, I introduced how to run a very simple HelloWorld script with Puppet 3.1.0 on CentOS 6.3. Although it shows how Puppet works, it’s not really how Puppet is used in real world. To get the most out of Puppet, you want to run the client/server mode where you have a master and many agents.
Part 1: Install Puppet server
Puppet is a very well-known configuration management tool that has been adopted by many enterprises and service providers. VMware recently invested $30M in Puppet Labs, the company behind it. It’s clear that VMware needs such a tool if it wants to grow in data center space.
I actually got a book from last year’s PuppetConf in San Francisco and browsed through it on my flights. As with any other technology, reading it does not mean getting it. To get my hands dirty, I played with it in my home lab last week.