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

[Glenn Satchell]

>Date: Thu, 29 May 2008 10:18:42 -0400
>To: dhcp-users at isc.org
>From: Michael Kaegler <Michael.Kaegler at marist.edu>
>Subject: Re: 3.1.1 Failover Speed (was: Re: 4.1.0a1 Failover sync speed)
>Cc: Chuck Anderson <cra at WPI.EDU>
>
>(the entire file is 3.7mb. The server goes from starting to serving 
>in less than 2 seconds normally.)
>
>>  > We do have 140 subnets (mostly /21s). The load averages on the test 
servers
>>  > were high (4.00 - 6.00) but the interfaces were only pushing 60kbit.
>>
>>140 * 2048 = 286,720.  I'm not sure what the expected performance
>>should be for synchronizing almost 300,000 leases over failover.
>
>I would expect failover transfers to be capable of running at the 
>same speed as regular DHCP requests.

Yeah, but the initial sync has to copy over 286,000 leases. You don't
normally hand aout 286k leases at the same time. Is the time similar on
a subsequent restart once the initial sync has comleted or does it
improve?

Have you tried snooping the network to see what traffic was being
transmitted?

>>  > There are fixed-address lines within those ranges which will be removed
>>  > (but can't be yet, we still need them).
>>
>>You have a configuration error.  You must not have fixed-addresses
>>that are within ranges.
>[...]
>>If you don't do this, you will confuse the dhcp server which will try
>>to assign dynamic IPs that match your fixed-address, causing an IP
>>conflict.
>
>This may have been true once, but we tested this: setup a subnet with 
>a "range 10.10.10.4 10.10.10.5" and assigned '.4 fixed. We joined two 
>machines, neither of which were the MAC assigned '.4, and no other 
>machines were on the subnet at all. The first was assigned .5, the 
>second just generated a 'no addresses available' error.

Umm, not necessarily. dhcpd sends a ping before offering the address to
make sure it is free. If it gets a response then it will abandon that
address and not attempt to re-use it until it has no other available
leases.

However, if the device with the fixed-address is switched off the
ping-check sees the address is available and allocates it. If the fixed
address device is now booted you wind up with duplicate IP addresses.
Maybe you have been lucky so far, but with that many hosts it is likely
to occur reasonably often. Likewise with any manually configured
devices that happen to be in the range.

Best practise says that you should not have fixed-address devices
inside your dynamic ranges.

regards,
-glenn