Re: lagg/lacp poor traffic distribution

2010-12-21 Thread Eugene Grosbein
On 21.12.2010 21:57, Fabien Thomas wrote: >> 1. Is it a bug or design problem? > > How many queues have you with igb? If it's one it will explain why the > flowid is bad for load balancing with lagg. How do I know? I've read igb(4) manual page and found no words >>> vmst

Re: lagg/lacp poor traffic distribution

2010-12-21 Thread Fabien Thomas
On Dec 21, 2010, at 3:48 PM, Eugene Grosbein wrote: > On 21.12.2010 20:41, Fabien Thomas wrote: > > 1. Is it a bug or design problem? How many queues have you with igb? If it's one it will explain why the flowid is bad for load balancing with lagg. >>> >>> How do I know? I'

Re: SOCK_STREAM socket in kernel space

2010-12-21 Thread Mohammad Hedayati
On Tue, Dec 21, 2010 at 4:23 PM, John Baldwin wrote: > On Tuesday, December 21, 2010 6:39:26 am Mohammad Hedayati wrote: >> I'm about to use a char device for a kind of distributed processing, >> so I've coded the open function as follows. The problem is that >> soaccept returns 0 without populati

Re: lagg/lacp poor traffic distribution

2010-12-21 Thread Eugene Grosbein
On 21.12.2010 20:41, Fabien Thomas wrote: 1. Is it a bug or design problem? >>> >>> How many queues have you with igb? If it's one it will explain why the >>> flowid is bad for load balancing with lagg. >> >> How do I know? I've read igb(4) manual page and found no words > vmstat -i will sho

Re: lagg/lacp poor traffic distribution

2010-12-21 Thread Fabien Thomas
On Dec 21, 2010, at 3:00 PM, Eugene Grosbein wrote: > On 21.12.2010 19:11, Fabien Thomas wrote: > I had this problem with igb driver, and I found, that lagg selects outgoing interface based on packet header flowid field if M_FLOWID field is set. And in the igb driver code flowid i

Re: lagg/lacp poor traffic distribution

2010-12-21 Thread Eugene Grosbein
On 21.12.2010 19:11, Fabien Thomas wrote: >>> I had this problem with igb driver, and I found, that lagg selects >>> outgoing interface based on packet header flowid field if M_FLOWID field >>> is set. And in the igb driver code flowid is set as >>> >>> #if __FreeBSD_version >= 80 >>> <--

Re: lagg/lacp poor traffic distribution

2010-12-21 Thread Fabien Thomas
>>> Hi! >>> >>> I've loaded router using two lagg interfaces in LACP mode. >>> lagg0 has IP address and two ports (em0 and em1) and carry untagged frames. >>> lagg1 has no IP address and has two ports (igb0 and igb1) and carry >>> about 1000 dot-q vlans with lots of hosts in each vlan. >>> >>> Fo

Re: SOCK_STREAM socket in kernel space

2010-12-21 Thread John Baldwin
On Tuesday, December 21, 2010 6:39:26 am Mohammad Hedayati wrote: > I'm about to use a char device for a kind of distributed processing, > so I've coded the open function as follows. The problem is that > soaccept returns 0 without populating the raddr. I've checked netstat, > everything seems to b

SOCK_STREAM socket in kernel space

2010-12-21 Thread Mohammad Hedayati
I'm about to use a char device for a kind of distributed processing, so I've coded the open function as follows. The problem is that soaccept returns 0 without populating the raddr. I've checked netstat, everything seems to be fine, the socket is created, bound and the state is LISTENING. Even the

Re: lagg/lacp poor traffic distribution

2010-12-21 Thread Eugene Grosbein
On 20.12.2010 17:21, Shtorm wrote: > On Sun, 2010-12-19 at 00:35 +0600, Eugene Grosbein wrote: >> Hi! >> >> I've loaded router using two lagg interfaces in LACP mode. >> lagg0 has IP address and two ports (em0 and em1) and carry untagged frames. >> lagg1 has no IP address and has two ports (igb0 an

Re: E4500 spend one core to saturate 1Gbit/s link with TCP -- is it nornal?

2010-12-21 Thread Lev Serebryakov
Hello, Eugene. You wrote 21 декабря 2010 г., 9:19:46: Is it normal, that 2.2GHz core is needed to saturate 1Gib link with only one client (and one TCP connction), or I have something misconfigured? >>> Compare with ftpd that uses sendfile() kernel function. >> simple "iperf"