addressalign-toparrow-leftarrow-rightbackbellblockcalendarcameraccwcheckchevron-downchevron-leftchevron-rightchevron-small-downchevron-small-leftchevron-small-rightchevron-small-upchevron-upcircle-with-checkcircle-with-crosscircle-with-pluscontroller-playcrossdots-three-verticaleditemptyheartexporteye-with-lineeyefacebookfolderfullheartglobegmailgooglegroupshelp-with-circleimageimagesinstagramFill 1launch-new-window--smalllight-bulblinklocation-pinm-swarmSearchmailmessagesminusmoremuplabelShape 3 + Rectangle 1ShapeoutlookpersonJoin Group on CardStartprice-ribbonprintShapeShapeShapeShapeImported LayersImported LayersImported Layersshieldstartickettrashtriangle-downtriangle-uptwitteruserwarningyahoo

Re: [TriangleDevOps] Monitoring Resourcees

From: user 3.
Sent on: Friday, August 24, 2012 8:21 AM
My information on Hyperic is a little dated but I'll share what I know. I worked at Red Hat on the Red Hat Network ("RHN") team when the company acquired JBoss. Part of that acquisition was a product called JBoss Operations Network ("JON") built on top of Hyperic. The product had many of the positive attributes you describe. It was easy to install, had a lot of plugins, and was relatively easy to operate. That said, performance and scalability were real problems. The agent was written in Java and required non-trivial resources on the managed server. The JON server was a monster in terms of CPU and memory and could only handle ~100 managed machines per JON server instance.

This information is quite dated by now and I'm sure both JON and Hyperic have improved. Given my past experiences with the product I would say I'm not surprised you're seeing resource usage and performance problems.

--Kevin
On Aug 23, 2012, at 8:21 PM, Greg Sheremeta wrote:

> What do people think of Hyperic? It can be kind of heavyweight compared to something written in C, and I've had problems with it flaking out on a few boxes. On a LAMP box I monitor, the HQ agent uses more memory and CPU than the apache server, which seems stupid. But it's a breeze to set up (especially compared to Nagios) and the plugin community is rich. For example, I had no problem getting it to work with my mongo cluster a few months back.
> 
> Greg
> 
> On Tue, Aug 21, 2012 at 12:06 AM, Mark Mzyk <[address removed]> wrote:
> This thread was What is DevOps, but it morphed into a monitoring thread. I'll return to what is DevOps later, but for now I wanted to offer up some monitoring resources for everyone.
> 
> Following up on what Craig said regarding the monitoring sucks project on github, this is the post that explains it all: http://lusislog.b...­ 
> 
> John Vincent, @lusis on Twitter, explains what made him start the repo and it's various purposes. The repo itself is here:
> 
> https://github.com...­
> 
> For great explanations on how Etsy uses monitoring, their Code as Craft blog is invaluable. Here is a blog post where they talk about using Graphite:
> 
> http://codeascraf...­
> 
> Follow their blog if you don't already.
> 
> Someone mentioned New Relic as a company that can provide monitoring for your app. Along with New Relic, BlueStripe is a local company that also provides a way to monitor your app. Boundary is another company doing interesting things in the space, as they provide network monitoring. I personally have not used any of these company's apps, but perhaps they might prove useful to you.
> 
> As far as doing monitoring you setup yourself, Nagios still appears to be king. Frequently I've seen Nagios paired with Munin or Cacti and more and more I'm seeing Graphite, sometimes paired with StatsD (Etsy blogged on Statsd here).
> 
> There is now also a tool called Rorschach to allow you to do alerting off of Graphite data.
> 
> Some of these things I've used, some of them I haven't. Clearly there is a lot going on in the monitoring space and it is evolving a lot after having stagnated for a while.
> 
> Coincidently, the next meetup is on Bronto's use of some of these monitoring tools. If you're interested in monitoring it would be a good meetup to attend.
> 
> - Mark 
> 
> On Wednesday, August 15, 2012 at 9:43 AM, Michael Hale wrote:
> 
>> Where I work (heroku.com) we use a bit of a home grown approach to metrics/monitoring/v­isibility using tools like graphite and some that we have open-sourced: http://vimeo.com/...­. One of the things I think is really cool is how we are able to alert on data that we trend to graphite. For example if service A has more than 100 connections over say 5 minutes we could trigger a nagios alert.
>> 
> 
> 
> 
> 
> 
> --
> Please Note: If you hit "REPLY", your message will be sent to everyone on this mailing list ([address removed])
> This message was sent by Mark Mzyk ([address removed]) from Triangle DevOps.
> To learn more about Mark Mzyk, visit his/her member profile
> Set my mailing list to email me As they are sent | In one daily email | Don't send me mailing list messages 
> 
> Meetup, PO Box 4668 #37895 New York, New York[masked] | [address removed]  
> 
> 
> 
> 
> 
> --
> Please Note: If you hit "REPLY", your message will be sent to everyone on this mailing list ([address removed])
> This message was sent by Greg Sheremeta ([address removed]) from Triangle DevOps.
> To learn more about Greg Sheremeta, visit his/her member profile
> Set my mailing list to email me As they are sent | In one daily email | Don't send me mailing list messages 
> 
> Meetup, PO Box 4668 #37895 New York, New York[masked] | [address removed]  

Our Sponsors

People in this
Meetup are also in:

Sign up

Meetup members, Log in

By clicking "Sign up" or "Sign up using Facebook", you confirm that you accept our Terms of Service & Privacy Policy