[Dnsmasq-discuss] Dhcp_release command casually fails when there are hundreds of dnsmasq instances

haixuan sun sunhaixuan at gmail.com
Wed Sep 7 07:18:40 BST 2016


Hi,

    We have a cloud(OpenStack) environment and created more than 200
private networks. Thus there are more than 200 dnsmasq instances in one
compute node.

    We find that dhcp_release command will casually fail. The success rate
is about 3/10.

    This causes the OpenStack cannot assign IP of a previously delete
instance.

    In another OpenStack environment, with less than 10 private network,
this problem does not happen.

    What is cause for this? Is that any socket buffer or queue, so that
dnsmasq randomly drop off the DHCP RELEASE packet?

    How to make sure the dhcp_release always send the DHCP RELEASE packet
to the server? Any configuration?

    dnsmasq version is 2.66.

-- 
Haixuan Sun
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.thekelleys.org.uk/pipermail/dnsmasq-discuss/attachments/20160907/2c134a08/attachment.html>


More information about the Dnsmasq-discuss mailing list