<div dir="ltr">After further investigation yesterday, I am not convinced it is an IP-address issue. The affected host machines are unable to start any existing or newly created containers. The incident that triggered the issue was cloning 1 container into 10 new ones, and then launching them all simultaneously. Are there any known concurrency issues with LXC which would explain why executing a lot of clone/start LXC commands at the same time causes all of LXC to be rendered useless until the host machine is physically rebooted?</div>
<div class="gmail_extra"><br><br><div class="gmail_quote">On Sat, Aug 10, 2013 at 11:42 AM, Tony Su <span dir="ltr"><<a href="mailto:tonysu@su-networking.com" target="_blank">tonysu@su-networking.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">FYI<br>
I avoid the whole issue assigning different IP addresses by creating<br>
my br devices using libvirt (vm manager).<br>
<br>
Using vm manager, I can<br>
Create a <named> virtual network that references using the specified<br>
bridge device (all libvrit created bridge devices have a "virbr" name<br>
instead of "br")<br>
When you setup your bridge device using libvirt, you can configure a<br>
DHCP process for that virtual network without having to install and<br>
run a proper DHCP server.<br>
<br>
Once you create the bridge device using libvirt, it can be used by<br>
<any> virtualization technology, so for example although I'm not<br>
managing my LXC Containers using libvirt, the bridge devices are still<br>
usable by KVM, Xen, LXC and QEMU on my machine. Once the bridge device<br>
is created, just reference it in your LXC Container config file and<br>
from within the container your eth0 if setup for DHCP will<br>
automatically get its address.<br>
<br>
You can display the bridge devices that exist on your machine with the<br>
following command<br>
brctl show<br>
<br>
Although it's probably likely that a "regular" bridge device could be<br>
configured with DHCP and even be referenced by name, I find it so much<br>
easier and avoids mistakes to just use vm manager to do the work for<br>
me.<br>
<br>
HTH,<br>
Tony<br>
<div><div class="h5"><br>
<br>
<br>
<br>
On Fri, Aug 9, 2013 at 7:32 PM, Serge Hallyn <<a href="mailto:serge.hallyn@ubuntu.com">serge.hallyn@ubuntu.com</a>> wrote:<br>
> Sorry, I can't figure out what's going wrong. You have unique macaddrs<br>
> for each container, so the dnsmasq-lxc should be handing out unique<br>
> ip addresses. What does /etc/network/interfaces in one of the containers<br>
> look like?<br>
><br>
>> ubuntu@ip-10-34-249-56:~$ lxc-version<br>
>> lxc version: 0.9.0<br>
><br>
> what about<br>
> dpkg -l | grep lxc<br>
> and<br>
> dpkg -l | grep dnsmasq<br>
> ?<br>
><br>
</div></div>> ------------------------------------------------------------------------------<br>
> Get 100% visibility into Java/.NET code with AppDynamics Lite!<br>
> It's a free troubleshooting tool designed for production.<br>
> Get down to code-level detail for bottlenecks, with <2% overhead.<br>
> Download for free and get started troubleshooting in minutes.<br>
> <a href="http://pubads.g.doubleclick.net/gampad/clk?id=48897031&iu=/4140/ostg.clktrk" target="_blank">http://pubads.g.doubleclick.net/gampad/clk?id=48897031&iu=/4140/ostg.clktrk</a><br>
> _______________________________________________<br>
> Lxc-users mailing list<br>
> <a href="mailto:Lxc-users@lists.sourceforge.net">Lxc-users@lists.sourceforge.net</a><br>
> <a href="https://lists.sourceforge.net/lists/listinfo/lxc-users" target="_blank">https://lists.sourceforge.net/lists/listinfo/lxc-users</a><br>
</blockquote></div><br></div>