This provides the ability to have a maximum number of workers, yet still allow scaling down. The scaling down is important to stop timeouts happening with under-utilised workers.
Note, although the code has workers_min, workers_max, the settings option --output-http-workers has remained and becomes the "max" value. I did this to hopefully not break too much for others. The min will default to 1, so users will see dynamic scaling unless they also set min to be the same. This will probably benefit them as they will avoid potential timeouts when the traffic reduces. But something to be aware of.
Added new `--http-pprof` option, example: "--http-pprof :8181"
It starts web server on given address, and expose special /debug/pprof
endpoint
See https://golang.org/pkg/net/http/pprof/
Added `—input-raw-buffer-size` - Controls size of the OS buffer (in
bytes) which holds packets until they dispatched. Default value depends
by system: in Linux around 2MB. If you see big package drop, increase
this value.
Additionally snaplen (max number of bytes being read for each packet)
now dynamically set based on interface MTU + max header size. In most
situations it should reduce package drop, because each packet will
consume less space in buffer.
Previously latency calcualted as Response.End - Request.Start
Where both End and Start is a last and first packets
This calcualtion is wrong, because it is total roundtrip
Correct server latency will be Response.End - Request.End
In addition added new `--input-raw-timestamp-type` option
which allows choose more precise packet timestamp source (if available).
Can be useful in case of non standard network interfaces when like
tunnels or SPAN ports, when IP of interface does not match with IP of
packet.
Can be used to read traffic from multiple ports and interfaces at once
as well.
The actual default is 5s, set in the call of NewHTTPClient,
even if we use `--output-http--timeout=0`.
It's clearer to user if this value id declared upfront.
Leonid Bugaev