Hi again,
I try to shape a 100 Mbit/s channel for FTP and UDP traffic, so one could not bother another.
I use PF's ALTQ with HSFC.
When I use these rules for 10 Mbit/s bandwidth, it works fine: if in FTP download process I start iperf, it works fine and shapes it so both run at approximately at 5 Mbit/s. But in a situation with 100 Mbit/s iperf just kills the FTP traffic. Is it iperf's specific or the same will be in real situation with UDP transmission software? Can I test it with some other software? (what software?)
I try to shape a 100 Mbit/s channel for FTP and UDP traffic, so one could not bother another.
I use PF's ALTQ with HSFC.
Code:
altq on em0 bandwidth 99Mb hfsc queue { std_up ftp_up sig_up }
queue std_up bandwidth 8% hfsc (default realtime 2% ecn )
queue ftp_up bandwidth 46% qlimit 500 hfsc (realtime 39% ecn)
queue sig_up bandwidth 46% qlimit 500 hfsc (realtime 39% ecn)
altq on em1 bandwidth 99Mb hfsc queue { std_down ftp_down sig_down }
queue std_down bandwidth 8% hfsc (default realtime 2% ecn )
queue ftp_down bandwidth 46% qlimit 500 hfsc (realtime 39% ecn)
queue sig_down bandwidth 46% qlimit 500 hfsc (realtime 39% ecn)When I use these rules for 10 Mbit/s bandwidth, it works fine: if in FTP download process I start iperf, it works fine and shapes it so both run at approximately at 5 Mbit/s. But in a situation with 100 Mbit/s iperf just kills the FTP traffic. Is it iperf's specific or the same will be in real situation with UDP transmission software? Can I test it with some other software? (what software?)