I think the issue might be that I'm running two VPP instances on the same
subnet and my upstream dhcp server is issuing the same address because of
client_id (DHCP option 61) being the same for both instances. I see option
61 added in 17.10 release
https://docs.fd.io/vpp/17.10/release_notes_1710.html & implemented here:
https://docs.fd.io/vpp/18.01/dd/de9/vnet_2dhcp_2client_8c.html#a697316f95ed097fc91c1dde48b61bc20

Is it added to startup configuration? I tried the following but didn't
work:

set dhcp client intfc <interface> [hostname <name>] [client_id <id>]

On Wed, Feb 21, 2018 at 5:34 PM, Matt Paska <[email protected]> wrote:

> I'm testing a basic vpp setup on ubuntu running vpp v18.01-release. In my
> config vpp receives dhcp and routes for other interfaces and a host tap
> interface. It works fine at initial vpp service start but after about an
> hour it stops. The service is active and I'm not changing any
> configurations but it requires a service restart to work again.
>
> I have vpp show run, show errors, config and startup.conf files here:
> https://gist.github.com/anonymous/e6952862b4e1d6984afa20b5ba1e5efe
>
> After a quick restart of the vpp service, everything is fine again:
> https://gist.github.com/anonymous/484dd31bf0a8f904e4416a04579980af
>
> Please let me know if I can provide more info.
>
> Thanks
> 
>
>

Reply via email to