3.1.1 Failover Speed (was: Re: 4.1.0a1 Failover sync speed)

Claus Holm Christensen iscdhcp at claushc.dk
Thu May 29 15:42:24 UTC 2008


Chuck Anderson skrev:
> That goes against every recommendation, documentation, and experience 
> I have ever seen.  Of course, the code is the final authority on this.  
> I haven't looked at the code myself, but I've seen the authors of the 
> code say that your experience isn't enforced by the code.

It's easy to test... Create random host statements for all but one IP 
address in a test scope, then attach two clients that don't match any of 
the random hosts. If both clients get an address, then there is no such 
enforcement.

This issue has haunted me a couple of times, and I know the log message 
from this problem quite well. The log message is only generated 
on-the-fly, when the DHCP server selects this IP for responding to a 
DISCOVER or REQUEST. That's certainly a nuisance!


-- 
Claus Holm Christensen


More information about the dhcp-users mailing list