dynip/statip IP overflow

[Oliver Hinckel]

Hello,

Gunther Mayer wrote on 06/01/2008 03:57 PM:
[...]
>> E.g. we have the following setup
>> dynip 192.168.174.128/25
>> statip 192.168.174.0/25
>>
>> In case a client will get the dynip address 192.168.174.128, the next 
>> client will get 192.168.175.1. It seems that the IP address is just 
>> increased by one without checking against the dynip network mask. This 
>> seems like an "overflow" too me.
> That makes no sense. According to the above 192.168.174.128 is neither 
> in the dynip range nor the last one.

.128 is the network number for the second network (like .0 for first 
network). You're right, that's not a real IP address which can be used.

> .129 would be the very first one 
> and .254 the last. Could you please give us a real example of "wrong" ip 
> address assignment?

Yes, .129 is the first IP address of this range. But I was in hurry and 
didn't realized that I have this typo in my mail. Anyway, thanks for 
noticing :)

> I'm operating a large network on v0.9.11 with almost identical setups 
> (also a statip .0/25 subnet with dynip .128/25), have experienced no 
> issues to date and in fact have just verified with one of the logfiles 
> of a busy site that indeed chilli wraps around correctly.

We'll check this again with a newer version of chilli as David suggested 
already.

Greetings
Oliver Hinckel