docs/index.asciidoc in logstash-codec-netflow-3.5.1 vs docs/index.asciidoc in logstash-codec-netflow-3.5.2

- old
+ new

@@ -40,14 +40,14 @@ |Cisco ASR | | y | | |Cisco IOS 12.x | | y | | |Cisco WLC | | y | | |Citrix Netscaler | | | y | Still some unknown fields, labeled netscalerUnknown<id> |fprobe | y | | | -|Fortigate FortiOS 5.2 | | y | | +|Fortigate FortiOS | | y | | |ipt_NETFLOW | y | y | y | |Juniper MX80 | y | | | SW > 12.3R8 -|Mikrotik 6.35.4 | y | | n | http://wiki.mikrotik.com/wiki/Manual:IP/Traffic_Flow +|Mikrotik | y | | y | http://wiki.mikrotik.com/wiki/Manual:IP/Traffic_Flow |nProbe | y | y | y | |OpenBSD pflow | y | n | y | http://man.openbsd.org/OpenBSD-current/man4/pflow.4 |Softflowd | y | y | y | IPFIX supported in https://github.com/djmdjm/softflowd |Streamcore Streamgroomer | | y | | |Ubiquiti Edgerouter X | | y | | With MPLS labels @@ -60,16 +60,35 @@ [source, ruby] -------------------------- input { udp { - port => 2055 + port => 2055 codec => netflow } } -------------------------- +For high-performance production environments the configuration below will decode up to 6000 flows/sec on an 8 CPU instance. If your total flowrate exceeds 6000 flows/sec, you should use multiple Logstash instances. + + +[source, ruby] +-------------------------- +input { + udp { + port => 2055 + receive_buffer_bytes => 16777216 + codec => netflow + workers => 6 + } +-------------------------- + +Make sure to increase the Linux kernel receive buffer limit: + + # sysctl -w net.core.rmem_max=$((1024*1024*16)) + + [id="plugins-{type}s-{plugin}-options"] ==== Netflow Codec Configuration Options [cols="<,<,<",options="header",] |======================================================================= @@ -105,10 +124,10 @@ ===== `cache_ttl` * Value type is <<number,number>> * Default value is `4000` -Netflow v9/v10 template cache TTL (minutes) +Netflow v9/v10 template cache TTL (seconds) [id="plugins-{type}s-{plugin}-include_flowset_id"] ===== `include_flowset_id` * Value type is <<boolean,boolean>>