Wow, that was fast!

Awesome, this is a huge help. Thank you.



On Wednesday, February 3, 2016 at 12:24:49 PM UTC-8, Santiago Bassett wrote:
>
> Hi Graeme,
>
> Victor implemented this yesterday in our fork: 
> https://github.com/wazuh/ossec-wazuh/commit/b277f0b159a0145d7501d446c429db19a50f922a
>
> It actually shows the wrong ID in the message (the one the agent is trying 
> to use):
>
> 2016/02/03 19:27:52 ossec-remoted(1408): ERROR: Invalid ID 1036 for the 
> source ip: XX.XX.XX.XX'
>
> I think he is working on a pull request.
>
> Best
>
> On Tue, Feb 2, 2016 at 10:19 AM, Graeme Stewart <[email protected] 
> <javascript:>> wrote:
>
>> Hi Santiago,
>>
>> Upon looking at the source, I'm not actually sure this is really a 
>> feasible ask. 
>>
>> Looks like the client sends a hash of it's ID and key to the manager and 
>> the manager then compares that to a table of ID/key hashes. So the actual 
>> client ID is never sent in the message in a way that would permit the 
>> manager to enumerate the true ID.
>>
>> 114 
>> <http://fossies.org/dox/ossec-hids-2.8.3/hash__op_8c.html#a59af72e305e0463ff084ce2bb41e5565>
>>  int _os_genhash 
>> <http://fossies.org/dox/ossec-hids-2.8.3/hash__op_8c.html#a59af72e305e0463ff084ce2bb41e5565>
>> (OSHash <http://fossies.org/dox/ossec-hids-2.8.3/struct__OSHash.html> *
>> self, char *key)
>> 115 {
>> 116  unsigned int hash_key = self->initial_seed;
>> 117 
>> 118  /* What we have here is a simple polynomial hash.
>> 119  * x0 * a^k-1 .. xk * a^k-k +1
>> 120  */
>> 121  while(*key)
>> 122  {
>> 123  hash_key *= self->constant;
>> 124  hash_key += *key;
>> 125  key++;
>> 126  }
>> 127 
>> 128  return(hash_key);
>> 129 }
>>
>> I might ask that the actual manager / ID failing hash be included in the 
>> error (or debug) log, that way we could "out of band" (not within OSSEC 
>> itself) attempt to identify client keys using an externalized key 
>> management process.
>>
>> Thanks for replying.
>>
>> Graeme
>>
>> On Tuesday, February 2, 2016 at 9:40:22 AM UTC-8, Santiago Bassett wrote:
>>>
>>> Hi Graeme,
>>>
>>> this is the message you refer to right?
>>>
>>> src/error_messages/error_messages.h:#define ENC_IP_ERROR    "%s(1408): 
>>> ERROR: *Invalid ID* for the source ip: '%s'."
>>>
>>> Feel free to open issues in github for these type of requests. Just 
>>> opened one for this one.
>>>
>>> Best regards
>>>
>>> On Fri, Jan 29, 2016 at 9:26 AM, Graeme Stewart <[email protected]> 
>>> wrote:
>>>
>>>> Would it really be difficult to actually show the error remote host ID 
>>>> in the ossec.log? This would make identifying key mismatch so much easier.
>>>>
>>>> -- 
>>>>
>>>> --- 
>>>> You received this message because you are subscribed to the Google 
>>>> Groups "ossec-list" group.
>>>> To unsubscribe from this group and stop receiving emails from it, send 
>>>> an email to [email protected].
>>>> For more options, visit https://groups.google.com/d/optout.
>>>>
>>>
>>> -- 
>>
>> --- 
>> You received this message because you are subscribed to the Google Groups 
>> "ossec-list" group.
>> To unsubscribe from this group and stop receiving emails from it, send an 
>> email to [email protected] <javascript:>.
>> For more options, visit https://groups.google.com/d/optout.
>>
>
>

-- 

--- 
You received this message because you are subscribed to the Google Groups 
"ossec-list" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to