That's the way I've ended up going as well, it definitely make managing 
extractors simpler since the extractors on an input all apply to the same 
types of messages.

The only gotcha I've run into is with testing filebeats, the collector 
allows you to set up and attempt to use multiple outputs however it seems 
that filebeats only supports one output so I've switched back to using 
nxlog again

On Thursday, 8 September 2016 03:11:28 UTC+10, 8bits...@gmail.com wrote:
>
> I use a different input for each type of log, platform, eventlog, iis, 
> etc..  My thinking was mainly I want to see everything from something 
> specific without noise from another and without the need for a stream.
>
> -
>
> On Wednesday, September 7, 2016 at 4:01:08 AM UTC-6, Michael Anthon wrote:
>>
>> While our system currently isn't that large I'm trying to determine the 
>> best way to configure Graylog to make future updates and extensions simple 
>> to manage.
>>
>> Where I'm struggling with this is with the impact in terms of performance 
>> of configuring things certain ways.
>>
>> So, for example, we have data being sourced from several different types 
>> of logs
>>
>>    - IIS Logs
>>    - nginx logs
>>    - Windows event logs
>>    - PHP Error logs
>>    - Custom application logs
>>    - syslogs from various devices and servers
>>    - tomcat/java logs
>>
>> Each of these different types has various requirements in terms of 
>> extractors and processing that we do to provide us with useful fields for 
>> searching.
>>
>> The options as I see them are 
>>
>>    1. create a small number of inputs that handle all the messages and 
>>    have a large set of extractors to deal with all the different message 
>> types 
>>    that come through the input.
>>    2. create an input for each type of message source with the 
>>    extractors for that type of message as needed
>>
>> To me, option 2 seems the more sensible in terms of future management and 
>> even initial setup but I'm unsure of the impact of having more inputs 
>> versus less inputs with more extractors.
>>
>> I'd appreciate any insight/advice on this (or pointers to documentation 
>> that I may have missed)
>>
>> Cheers,
>> Michael
>>
>

-- 
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 graylog2+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/graylog2/87a77fa3-4a61-42c2-8170-329050e15a83%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to