On 22.8.2012. 20:50, Florian Obser wrote:
> Hi,
> 
> I think I got this now.
> - replace time_seconds with time_uptime
> - with that flow_finish in pflow can be simplified (pointed out by benno@)
>   this should take care of flows with finish < start for localy created
>   states
> - change various variables from unsigned to signed (mainly) in pfsync which
>   could underflow (inspired by a patch from dlg@ from last year, pointed out
>   by cameild@). This fixes pflow and at the same time an underflow problem
>   in pfsync which dlg's patch was addressing. If I understand the history
>   correctly cameild@ noticed this problem last year.
> 
> Hrvoje Popovski tested an older version without the pfsync fix.
> 
> Since this changes the semantics of a field in pfsync both pfsync machines
> need to be updated (I think it's ok to import 5.1 states into this version
> but the other way around will cause problems.)
> 
> We are running with this patch since yesterday on a redundant firewall
> (amd64) with pfsync and no longer see broken flows (i.e. where finish
> < start). The pair survived various failovers.
> 


Hello,

I can confirm that there aren't any 4294* flow durations in nfdump
(1.6.6). Pfsync is patched and enabled on firewalls and redundancy was
tested.

Thank you.

Reply via email to