With Chef, you write abstract definitions as source code to describe how you want each part of your infrastructure to be built, and then apply those descriptions to individual servers.
The result is a fully automated infrastructure: when a new server comes on line, the only thing you have to do is tell Chef what role it should play in your architecture.
Chef allows you to quickly deploy or spin servers up and down. When deploying a server, you can set tags and labels, so when the server is available, the Server Monitoring widget that represents the servers is immediately populated with the meta data to allow for more efficient and targeted filtering and alerting.
CopperEgg Server Monitoring monitors critical OS, system, and process statistics – in real-time – across Linux servers in any cloud environment.
In addition, Server Monitoring real-time monitoring supports Windows, FreeBSD, MacOS X, and most other 2.6+ Linux distributions, including Debian, RHEL, CentOS, Fedora, Amazon, SuSE, openSuSE, Vyatta, and Gentoo).
Critical Linux performance monitoring data collected includes:
• CPU utilization, per CPU and in-aggregate
• Details of the individual components of CPU utilization, including CPU steal and I/O wait
• Memory stats like cache, buffer, active
• Network TX and RX across each NIC and in-aggregate
• Disk volume consumption by volume
• Disk I/O
• Swap Activity
• Processes, CPU, and memory (like top on steroids)
• And many more.
3 ways process monitoring with CopperEgg can help you:
1. Easily identify a process that has consumed too much memory or CPU. If a system is running high on CPU or memory, you can detect which processes may need to be stopped or restarted.
2. Discover if an issue arises by 2 or more processes or jobs running simultaneously. This might cause a high load on the system and may extend the run time of the combined jobs as opposed to them running at various times. Example: running an AV scan while trying to do a backup.
3. Gain visibility into multiple CPUs to determine if a process or job may not be written to take advantage of multiple CPUs. If this was not visible, a job may take longer to execute than anticipated. With CopperEgg Server Monitoring you can verify that a job is only using one CPU as opposed to multiple.
The CopperEgg Chef recipe is available at https://github.com/CopperEgg/chef-copperegg.
• Chef overview: http://wiki.opscode.com/pages/viewpage.action?pageId=7274862
• Chef news: http://www.opscode.com/news-archive/
• Blog: http://copperegg.com/hail-to-the-chef/
• Get the recipe: https://github.com/CopperEgg/chef-copperegg
• Impact of Real-time on system performance: http://copperegg.com/real-time-stats-collection-and-monitoring-with-negligible-system-impact/
• GeekAustin DevOps preso: http://www.slideshare.net/mattray/geekaustin-devops