Hi,

please don't "hijack" old mailing list threads. You can always open a new 
one…

Currently there's pretty much only the 
org.graylog2.filters.ExtractorFilter.executionTime metric (see 
http://localhost:12900/system/metrics/org.graylog2.filters.ExtractorFilter.executionTime?pretty=true)
 
which hints to how long all extractors have been running. This isn't very 
detailed, though.


I must wait perhaps 1 hour, to display the older messages hour hour ago


This sounds like a simple timezone issue. Please check that the account 
you're using the web interface with is using the correct timezone (System 
-> Users); in case of the admin user, take a look at the root_timezone 
setting in the Graylog configuration file (
https://github.com/Graylog2/graylog2-server/blob/1.1.6/misc/graylog2.conf#L28-L31).
 
Additionally, if message don't contain a proper timestamp (i. e. including 
a timezone), UTC is being used by default.


Cheers,
Jochen

On Monday, 24 August 2015 09:54:45 UTC+2, HappyDaysInfo wrote:
>
> Hi Jamie, 
>
> For information , and concerning my first experience using graylog, 
>
> I 've defined an input udp syslog with haproxy extractirs (and I have 
> checked all extrators). 
> An haproxy Aloha forward messages in 5514 ports and I have noticied , 
> something strange:
>
> I must wait perhaps 1 hour, to display the older messages hour hour 
> ago(input receveid messages and search modukle)   parsed and with the only 
> condition if I disable the input before
> If I don't use extractorn the messages are displayed in real time.
>
> In my architecture I have separated elastic search in a different 
> equipement than graylor server and it's web interface.
>
> I have take  look in index performance from elastic search node (one 
> shark), but i don't notice problem performance.
>
> My query is: is there a log activity concerning parsing extractor, because 
> there is no doubt the impacte performance, but I don't understand How it 
> can impacte the avaibility of consulting the messages already loggued
>
> Best regards.
>
>  
>
> Le mercredi 24 septembre 2014 20:10:05 UTC+2, Jochen Schalanda a écrit :
>>
>> Hi Jamie,
>>
>> Am Mittwoch, 24. September 2014 19:58:38 UTC+2 schrieb Jamie Geyer:
>>>
>>> just a quick question -
>>> In general does adding extractors to an input cause any performance 
>>> impacts on the input?
>>>
>>
>> In general yes, because the extractors have to be run for every message 
>> which is processed by that input. The concrete impact varies between 
>> "negligible" and "a lot" depending on the number and complexity of 
>> extractors.
>>
>> I recommend taking a look at the stream_processing_timeout and 
>> stream_processing_max_faults settings in your graylog2.conf and 
>> configure them accordingly (or keep the quite sane defaults).
>>
>>
>> Cheers,
>> Jochen
>>
>>

-- 
You received this message because you are subscribed to the Google Groups 
"Graylog Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/graylog2/9630cb9a-ff38-44d1-8d53-76c72ea3643a%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to