[Lxc-users] Monitoring per container

[David Parks]

Does anyone have any pointers on how I might monitor things like CPU and
DISK activity PER CONTAINER? (Ubuntu 12.10 server here)

 

I saw something on You Tube using RHL that demoed it beautifully, but I'm
looking for something a bit more rudimentary, maybe that I could plug into
Nagios or use to just see how things look under load.  

 

Running top-like utilities on the host doesn't really split it up well
enough by container, even htop with cgroups is difficult at best. 

 

I really would love a view like this on the host, like an htop view grouped
by cgroup: 

 

Container            cpu         IO                           network

CN1                        2.5%      3Mb/sec              1Mbit

CN2                        30.1%    30Mb/sec            7.2Mbit

 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxcontainers.org/pipermail/lxc-users/attachments/20130511/a34f96eb/attachment.html>