Hi Raoul
you should be able to run npcapextract on your network storage
as long as npcapextract is able to access both the timeline and the pcap files
and the timeline links have correct paths (the timeline is a tree of links to 
pcap and index files).
BTW, are you aware of the new ‘nxn' interface? This is a new GUI to run 
extractions, included in the nBox package. Please take a look at 
As of flushing data to disk, you have those 2 options you listed below, the 
one requires less efforts on your side as there is no need to broadcast a signal
to all the appliances, with the drawback that you can run queries on 10-min old
data (with the default configuration) or N-secs old data is you configure a max
duration (this depends on your use case).


> On 10 Apr 2018, at 22:45, Raoul Duke <rduke...@gmail.com> wrote:
> Hi,
> Is it viable / recommended to write n2disk cache to (low latency) network 
> storage?
> I have multiple servers I would like to capture certain traffic profile on 
> but then some time later be able to extract pcaps realated to certain 
> application events.   My thinking is I'd rather not have to do the 
> npacpextract "on box" and then have to work out how to copy them somewhere 
> else on the network when I could just do the npacpextract directly off 
> network storage.
> Are there any limitations / gotchas I should be aware of if using 
> npacpextract from a network attached cache while it is being written?
> Also, this comes back to a question I asked the other day.  but assuming I 
> wanted to npacpextract a certain pcap that happened between timestamps X and 
> Y on port Z but I'm not sure yet if that data has been yet flushed to disk.  
> are my only options either to:
> * use the [--max-file-duration|-t] $secs option and wait $secs after the 
> associated application event until attempting to npacpextract
> * force an explicit flush by sending a USR1 signal to the n2disk process. 
> (this approach would be complicated if the cache is on network storage and 
> the consumer wishing to run npacpextract is on a different machine than the 
> n2disk process)
> It seems like the first option is the least worst approach.  Can you suggest 
> any others?
> Thanks,
> RD
> _______________________________________________
> Ntop mailing list
> Ntop@listgateway.unipi.it
> http://listgateway.unipi.it/mailman/listinfo/ntop

Attachment: signature.asc
Description: Message signed with OpenPGP

Ntop mailing list

Reply via email to