Please file a bug on github

Luca

On 12/14/2015 02:16 PM, Munroe Sollog wrote:
> It would be great to have the option to "Bring-Your-Own-Indexer" so we can 
> use some thing like
> logstash to pull from redis and manipulate fields as necessary.
>
> - Munroe
>
> On 12/13/2015 10:06 AM, Simone Mainardi wrote:
>> Hi, please see below inserted comments
>>
>> simone
>>
>> On Thu, Dec 10, 2015 at 7:31 PM, Munroe Sollog <[email protected] 
>> <mailto:[email protected]>> wrote:
>>
>>     I'm pushing netflow into elasticSearch and it looks like there is a 
>> block of fields that come in as
>>     json_N.  I've included a sample document from kibana.  Two questions:
>>
>>     1) Is there a way I can push these docs into redis instead so that I can 
>> then pull it out using
>>     logstash so I have the ability to mutate fields?
>>
>>
>> not directly. ntopng pushes flows into redis before exporting them to ES, 
>> but the redis queue is
>> only accessible to the ntopng internals.
>>  
>>
>>     2) Is there a way I can define these fields before ntopng ships them to 
>> elasticSearch?
>>
>>
>> these fields are hard-coded. you have to edit the ParserInterface.cpp file 
>> and recompile ntopng to
>> customize them.
>>  
>>
>>
>>
>>     here's the json:
>>      {
>>       "_index": "ntopng-2015.12.10",
>>       "_type": "ntopng",
>>       "_id": "AVGNJSTcITc7jbmnrBAl",
>>       "_score": null,
>>       "_source": {
>>         "@timestamp": "2015-12-10T18:26:04.0Z",
>>         "type": "ntopng",
>>         "IPV4_SRC_ADDR": "192.168.118.16",
>>         "L4_SRC_PORT": 52009,
>>         "IPV4_DST_ADDR": "199.16.156.70",
>>         "L4_DST_PORT": 443,
>>         "PROTOCOL": 6,
>>         "L7_PROTO": 91,
>>         "L7_PROTO_NAME": "SSL",
>>         "TCP_FLAGS": 0,
>>         "IN_PKTS": 8,
>>         "IN_BYTES": 838,
>>         "OUT_PKTS": 0,
>>         "OUT_BYTES": 0,
>>         "FIRST_SWITCHED": 1449771964,
>>         "LAST_SWITCHED": 1449771964,
>>        ##HERE IS THE BLOCK OF WEIRD FIELDS##
>>         "json": {
>>           "5": "0",
>>           "9": "0",
>>           "10": "1",
>>           "13": "0",
>>           "14": "16",
>>           "15": "0.0.0.0",
>>           "16": "6522",
>>           "17": "13414",
>>           "42": "32102093"
>>         },
>>       ##END OF WEIRD FIELDS##
>>
>>
>> These `weird` fields are actually netflow Field Type Definitions that are 
>> not explicitly parsed by
>> ntopng. In order not to lose information, these fields are collected into a 
>> valid "json" field of
>> the exported data.
>> (see 
>> http://www.cisco.com/en/US/technologies/tk648/tk362/technologies_white_paper09186a00800a3db9.html)
>>  
>>
>>         "CLIENT_NW_LATENCY_MS": 0,
>>         "SERVER_NW_LATENCY_MS": 0,
>>         "SRC_IP_COUNTRY": "US",
>>         "SRC_IP_LOCATION": [
>>           -75.354698,
>>           40.590199
>>         ],
>>         "DST_IP_COUNTRY": "US",
>>         "DST_IP_LOCATION": [
>>           -122.393303,
>>           37.769699
>>         ],
>>         "PASS_VERDICT": true
>>       },
>>       "fields": {
>>         "@timestamp": [
>>           1449771964000
>>         ]
>>       },
>>       "sort": [
>>         1449771964000
>>       ]
>>     }
>>
>>     --
>>     Munroe Sollog
>>     LTS - Network Analyst
>>     x85002
>>     _______________________________________________
>>     Ntop mailing list
>>     [email protected] <mailto:[email protected]>
>>     http://listgateway.unipi.it/mailman/listinfo/ntop
>>
>>
>>
>>
>> _______________________________________________
>> Ntop mailing list
>> [email protected]
>> http://listgateway.unipi.it/mailman/listinfo/ntop
>>

_______________________________________________
Ntop mailing list
[email protected]
http://listgateway.unipi.it/mailman/listinfo/ntop

Reply via email to