Re: connbytes & 64bit counters

[Patrick McHardy <kaber@trash.net>]

Krzysztof Oledzki wrote:
> On Wed, 10 Jan 2007, Patrick McHardy wrote:
> 
>>> linux-2.6.19.1-64bitconntrack/net/ipv4/netfilter/ip_conntrack_core.c
>>> 2006-12-22 21:13:35.000000000 +0100
>>> @@ -1148,9 +1148,6 @@
>>>          ct->counters[CTINFO2DIR(ctinfo)].packets++;
>>>          ct->counters[CTINFO2DIR(ctinfo)].bytes +=
>>>                          ntohs(skb->nh.iph->tot_len);
>>> -        if ((ct->counters[CTINFO2DIR(ctinfo)].packets & 0x80000000)
>>> -            || (ct->counters[CTINFO2DIR(ctinfo)].bytes & 0x80000000))
>>> -            event |= IPCT_COUNTER_FILLING;
>>
>>
>>
>> This was actually broken before, since the counters are not
>> reset (they just overflow) an event was generated for every
>> packet until the overflow once they reached 2^31. Anyway,
>> I'm not sure how ulogd2 uses these counters, Harald, is it
>> necessary to receive period updates?
> 
> 
> If we remove IPCT_COUNTER_FILLING there is no other solution than also
> drop this part of the code.

Let's wait what Harald says about ulogd2, I'm not sure whether we need
to keep this.

> BTW: what about CTA_COUNTERS32_*? Should we also drop that attributes?

They should be kept to make sure the value are not reused. Please just
add a comment stating that they are not used anymore.