config help - scaling problem
glenn.satchell at uniq.com.au
Thu Mar 4 11:13:58 UTC 2010
On 03/03/10 11:09, Jerimiah Cole wrote:
> Marc Perea wrote:
>> We've had a lot of problems in years past with dhcp protocol failover,
>> mostly communications interrupted and all leases owned by peer, so
>> we're currently using a warm standby with frequent backups of the
>> .conf. As noted, we really don't care about the actual state of the
>> lease, so we don't even really need the .leases file at all except for
>> the operation of the server.
> Me too. The strategy is to monitor carefully and put processes in place
> to respond before service degrades. You'd have to do this anyway with
> failover in place. We chose to avoid the additional complexity of failover.
At a former job (about 2004-2008) we ran dhcpd 3.0.5 (I think) with
failover for 2000 PCs and 4000 phones. We had relatively few failover
issues, and due to careful monitoring we were able to get on top of the
very infrequent problems. Mostly a restart of one dhcpd fixed the
problem, I only recall having to remove a lease file once I think. As
far as I know it is still running :)
If you don't use failover, then you use some other type of complexity,
such as a hot standby server and something like VRRP, or take on the
risk of a single server. All three are valid options depending on your
skills, site policy, requirements, and so on.
>> Since all of our environment comes from the .conf and not the .leases,
>> does anyone see any reason why we shouldn't have 1 or more other
>> servers also performing active dhcp servicing (as was recommended)? If
>> I were to point my relays to several DHCP servers, it seems I'd have
>> more redundancy and failover without having to use any protocol, which
>> is a benefit in my mind. Any flaws in that thinking?
If you're using hosts with fixed-address statements and no dynamic
ranges, then you can have as many servers with identical config as you
like. This is because the answer to a request is always the same IP
address; lease length doesn't really matter here. In this situation no
failover configuration is needed.
More information about the dhcp-users