I only have a few types of really high volume logging, its udp syslog from network devices. Havent tried other inputs at such high message volumes.
/Martin On Friday, 14 February 2014 14:15:48 UTC+1, GambitK wrote: > > Do you have experience with other, "non-udp" inputs? > I tried the TCP one and seemed fast. > > El viernes, 14 de febrero de 2014 05:08:23 UTC-4, Martin René Mortensen > escribió: >> >> Im with you, just be aware of its shortcomings and you will do fine. I >> recommend you keep your logging architecture, its good and effective for >> now. >> >> I dont know details about archiving - I archive to /dev/null myself - but >> you could dump an index in any format you wish with some lines of code in >> your favourite language - or even with a few well constructed curl commands. >> >> There is a feature ticket in graylog2 for using Grok filters, which is >> one of the strenghts in logstash - if that gets implemented it should get >> much easier to do field extractions and your need for logstash might >> disappear - I just hope performance wont suffer overly much. >> >> On Friday, 14 February 2014 01:23:11 UTC+1, GambitK wrote: >>> >>> I like logstash flexibility in input, oputput types and powerful data >>> manipulation capabilities. >> >> -- You received this message because you are subscribed to the Google Groups "graylog2" 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/groups/opt_out.
