Yann Berthier wrote:

>   Hello,
>
>On Thu, 02 Aug 2007, at 17:28, Maurizio Molina wrote:
>
>  
>
>>But the active timeout in my routers is 300 sec!!, so I would expect to
>>see a persistent flow to be exported, at most, 1 time every 5 minutes.
>>Also, it cannot be due to the inactive timeout (which in my routers is
>>60 s).
>>    
>>
>
>   I don't understand your point - it looks like your flows are exported
>   roughly every 300 s (see the duration)
>  
>
I'm puzzled because every 300s I get (roughly) 10 more flows with this
same 5-tuple (see the reported number of aggregated flows: 10,20, 28,
35, ....). So I guess that this is the result of 10 separate exports
(why???), than are then aggregated into one flow. The duration of the
*aggregated* flow is probably referred to the difference between the
first and last packet of all these flows, and then it's reasonable that
it increases by 300 s every 300 s.

>   Is your router a Juniper ?
>
>  
>
yes - Junos release 8.2
Here's the part relevant for the Netflow configuration.

sampling {
    traceoptions {
        file sampled.trace size 3m;
    }
    input {
        family inet {
            rate 1000;
        }
    }
    output {
        cflowd X.Y.W.Z {
            port 7712;
            version 5;
            no-local-dump;
            autonomous-system-type peer;
        }
        flow-active-timeout 300;
    }
}

As you see, active timeout is configured to 300s, Inactive timeout is
not explicitly configured, than it's teh default (60s), according to
http://www.juniper.net/techpubs/software/junos/junos82/swconfig82-services/html/flow-monitoring-summary29.html#1055868

Maurizio

>>Any hint? CAn anybody check, by identifying a persistent flow their
>>network, if they observe a behavior consistent with their active timeout?
>>
>>Date flow start          Duration Proto      Src IP Addr:Port         
>>Dst IP Addr:Port   Packets    Bytes Flows
>>5 min slot
>>2007-07-26 22:19:46.432   270.910 TCP       x.138.4.19:22    -> 
>>y.220.165.210:48117      580   806816    10
>>10 min slot
>>2007-07-26 22:19:46.432   571.423 TCP       x.138.4.19:22    -> 
>>y.220.165.210:48117     1013  1410084    20
>>15 min slot
>>2007-07-26 22:19:46.432   814.342 TCP       x.138.4.19:22    -> 
>>y.220.165.210:48117     1469  2048588    28
>>20 min slot
>>2007-07-26 22:19:46.432  1160.044 TCP       x.138.4.19:22    -> 
>>y.220.165.210:48117     2099  2935420    35
>>......
>>60 min slot
>>2007-07-26 22:19:46.432  3521.209 TCP       x.138.4.19:22    -> 
>>y.220.165.210:48117     6048  8441088   112
>>
>>
>>-------------------------------------------------------------------------
>>This SF.net email is sponsored by: Splunk Inc.
>>Still grepping through log files to find problems?  Stop.
>>Now Search log events and configuration files using AJAX and a browser.
>>Download your FREE copy of Splunk now >>  http://get.splunk.com/
>>_______________________________________________
>>Nfsen-discuss mailing list
>>[email protected]
>>https://lists.sourceforge.net/lists/listinfo/nfsen-discuss
>>    
>>
>
>  
>


-- 
______________________________________________________________________

Maurizio Molina
Network Engineer

DANTE - www.dante.net

Tel: +44 (0)1223 371 300
Fax: +44 (0)1223 371 371
Email: [EMAIL PROTECTED]
PGP Key ID: 3FF58D51

City House, 126-130 Hills Road  
Cambridge CB2 1PQ                        
UK
_____________________________________________________________________



-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
Nfsen-discuss mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/nfsen-discuss

Reply via email to