Re: A problem on TCP in High RTT Environment.

On 2014-8-12, at 12:31, Michael Tuexen wrote: > On 12 Aug 2014, at 10:02, Eggert, Lars wrote: >> I bumped kern.ipc.nmbclusters by a factor of 100 (from 2036224 to >> 203622400). As Hiren said, kern.ipc.nmbufs auto-adjusted (from 13031835 to >> 205111860). > Just to double check: You changed it

Re: A problem on TCP in High RTT Environment.

On 12 Aug 2014, at 10:02, Eggert, Lars wrote: > Hi, > > On 2014-8-12, at 1:52, hiren panchasara wrote: >> On Mon, Aug 11, 2014 at 12:59 PM, Michael Tuexen >> wrote: >>> If I remember correctly, I increased >>> kern.ipc.nmbufs and kern.ipc.nmbclusters in /boot/loader.conf >> >> I believe, you

Re: A problem on TCP in High RTT Environment.

I use a switch and capture in the a sender mirror port. and I also noticed that some acks are before segment. I am not sure how to solve the problem. But, for my kvm-based virtual machines experimental environment. These are no such issues. ​ testtest.tar.gz

Re: A problem on TCP in High RTT Environment.

Hi, On 2014-8-12, at 1:52, hiren panchasara wrote: > On Mon, Aug 11, 2014 at 12:59 PM, Michael Tuexen > wrote: >> If I remember correctly, I increased >> kern.ipc.nmbufs and kern.ipc.nmbclusters in /boot/loader.conf > > I believe, you just need to set kern.ipc.nmbclusters (max mbuf > clusters a

Re: A problem on TCP in High RTT Environment.

On Mon, Aug 11, 2014 at 12:59 PM, Michael Tuexen wrote: > On 11 Aug 2014, at 21:40, Eggert, Lars wrote: > >> Hi, >> >> On 2014-8-11, at 21:27, Michael Tuexen >> wrote: >>> On 11 Aug 2014, at 14:12, Eggert, Lars wrote: 544/57/8194 requests for mbufs denied (mbufs/clusters/mbuf+clusters) >>

Re: A problem on TCP in High RTT Environment.

On 11 Aug 2014, at 21:40, Eggert, Lars wrote: > Hi, > > On 2014-8-11, at 21:27, Michael Tuexen > wrote: >> On 11 Aug 2014, at 14:12, Eggert, Lars wrote: >>> 544/57/8194 requests for mbufs denied (mbufs/clusters/mbuf+clusters) >> I guess the above is the problem. The card wants a lot of mbufs.

Re: A problem on TCP in High RTT Environment.

Hi, On 2014-8-11, at 21:27, Michael Tuexen wrote: > On 11 Aug 2014, at 14:12, Eggert, Lars wrote: >> 544/57/8194 requests for mbufs denied (mbufs/clusters/mbuf+clusters) > I guess the above is the problem. The card wants a lot of mbufs... > So the problem should go away if you increase the numbe

Re: A problem on TCP in High RTT Environment.

On 11 Aug 2014, at 14:12, Eggert, Lars wrote: > On 2014-8-11, at 9:17, Michael Tuexen > wrote: >> Was there any suspicious output provided by netstat -m when the problems >> occur? > > root@laurel:~ # netstat -m > 8186/2179/10365 mbufs in use (current/cache/total) > 8184/1214/9398/2036224 mbu

Re: A problem on TCP in High RTT Environment.

Niu Zhixiong wrote this message on Sun, Aug 10, 2014 at 20:27 +0800: > Hi, I am not sure whether my last email is filtered by mailing list. > After disabled tso??? the speed become even poorer??? > This is the packets captures. Plz see google drive. > tcp_with_tso_off.pcapng.gz >

Re: A problem on TCP in High RTT Environment.

For 60s test at 20Mbps, 200ms delay with hw.igb.num_queues=2 speed is 9.5Mbps. and without it the speed is 9.1Mbps. However, SCTP is 17.10. For 60s test at 20Mbps, 150ms delay with hw.igb.num_queues=2 speed is 11.1Mbps. and without it the speed is 9.92Mbps. However, SCTP is 18.59. For 60s test a

Re: A problem on TCP in High RTT Environment.

On 2014-8-11, at 9:17, Michael Tuexen wrote: > Was there any suspicious output provided by netstat -m when the problems > occur? root@laurel:~ # netstat -m 8186/2179/10365 mbufs in use (current/cache/total) 8184/1214/9398/2036224 mbuf clusters in use (current/cache/total/max) 8184/885 mbuf+clust

Re: A problem on TCP in High RTT Environment.

On 11 Aug 2014, at 09:32, Niu Zhixiong wrote: > Thanks for your reminding. I tried hw.igb.num_queues=2 just now. But, the > throughput is still slow. And When I tested same things in my Virtual > machine-based environment(use Virtio), the throughput is similar(SCTP is 2x > than TCP). So is the

Re: A problem on TCP in High RTT Environment.

On 11 Aug 2014, at 09:31, Niu Zhixiong wrote: > During TCP benchmark. > > root@Freetest0: # netstat -m > 35806/2129/37935 mbufs in use (current/cache/total) > 35251/1037/36288/1011638 mbuf clusters in use (current/cache/total/max) > 35251/1030 mbuf+clusters out of packet secondary zone in use (c

Re: A problem on TCP in High RTT Environment.

During TCP benchmark. root@Freetest0: # netstat -m 35806/2129/37935 mbufs in use (current/cache/total) 35251/1037/36288/1011638 mbuf clusters in use (current/cache/total/max) 35251/1030 mbuf+clusters out of packet secondary zone in use (current/cache) 0/6/6/505818 4k (page size) jumbo clusters in

Re: A problem on TCP in High RTT Environment.

Thanks for your reminding. I tried hw.igb.num_queues=2 just now. But, the throughput is still slow. And When I tested same things in my Virtual machine-based environment(use Virtio), the throughput is similar(SCTP is 2x than TCP). Regards, Niu Zhixiong －－－ kaia...@gmail.com On Mon,

Re: A problem on TCP in High RTT Environment.

On 11 Aug 2014, at 09:08, Eggert, Lars wrote: > Hi, > > On 2014-8-10, at 5:48, Niu Zhixiong wrote: >> I am using Intel I350-T4 NIC. > > igb driver? > > I've been having weird issues with this driver under 10-RELEASE, too. On one > machine, I had to limit hw.igb.num_queues=2 in order to get a

Re: A problem on TCP in High RTT Environment.

Hi, On 2014-8-10, at 5:48, Niu Zhixiong wrote: > I am using Intel I350-T4 NIC. igb driver? I've been having weird issues with this driver under 10-RELEASE, too. On one machine, I had to limit hw.igb.num_queues=2 in order to get any sort of useful connectivity. On another machine, I had to sev

Re: A problem on TCP in High RTT Environment.

Hi, I am not sure whether my last email is filtered by mailing list. After disabled tso， the speed become even poorer． This is the packets captures. Plz see google drive. tcp_with_tso_off.pcapng.gz Regards, Niu Zhixi

Re: A problem on TCP in High RTT Environment.

On 08/10/2014 12:23, John-Mark Gurney wrote: [..] The next thing would be to get a tcpdump, and take a look at the window size.. Wireshark has lots of neat tools to make this analysis easy... Another tool that is good is tcptrace.. It can output a variety of different graphs that wil

Re: A problem on TCP in High RTT Environment.

Niu Zhixiong wrote this message on Sun, Aug 10, 2014 at 11:56 +0800: > Actually. In the > http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/configtuning-kernel-limits.html > 12.11.2.2. TCP Bandwidth Delay Product > I saw an option called > net.inet.tcp.inflight.enable > net.inet.tcp.infligh

Re: A problem on TCP in High RTT Environment.

Niu Zhixiong wrote this message on Sun, Aug 10, 2014 at 11:48 +0800: > I am using Intel I350-T4 NIC. The LRO is closed by default. And by the way, > when I am using KVM-based virtual machine(virtio NIC) do the exactly same > test. The results are same. Have you tried disabling tso? I asked that i

Re: A problem on TCP in High RTT Environment.

Actually. In the http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/configtuning-kernel-limits.html 12.11.2.2. TCP Bandwidth Delay Product I saw an option called net.inet.tcp.inflight.enable net.inet.tcp.inflight.debug net.inet.tcp.inflight.min But, in FreeBSD 9.3R and 10R. I cannot find an

Re: A problem on TCP in High RTT Environment.

I am using Intel I350-T4 NIC. The LRO is closed by default. And by the way, when I am using KVM-based virtual machine(virtio NIC) do the exactly same test. The results are same. ifconfig igb0 igb0: flags=8843 metric 0 mtu 1500 options=403bb ether a0:36:9f:38:27:d0 inet 10.0.10.3 netmask 0xff0

Re: A problem on TCP in High RTT Environment.

- Original Message - From: "John-Mark Gurney" Niu Zhixiong wrote this message on Sun, Aug 10, 2014 at 10:50 +0800: I am sorry that I upload a WRONG SCTP capture. But, the throughput is same. SCTP is double than TCP, about 18Mbps. ??? sctp_2.pcapng.gz

Re: A problem on TCP in High RTT Environment.

Niu Zhixiong wrote this message on Sun, Aug 10, 2014 at 10:50 +0800: > I am sorry that I upload a WRONG SCTP capture. But, the throughput is same. > SCTP is double than TCP, about 18Mbps. > ??? > sctp_2.pcapng.gz > >

Re: A problem on TCP in High RTT Environment.

I am sorry that I upload a WRONG SCTP capture. But, the throughput is same. SCTP is double than TCP, about 18Mbps. ​ sctp_2.pcapng.gz ​ Regards, Niu Zhixiong －－－ kaia...@gmail.com On Sun, Aug 10, 2014

Re: A problem on TCP in High RTT Environment.

I am sure that wnd is about 2MB all the time. This is my latest capture, plz see Google Drive. In the latest test, TCP(0s-120s) is about 9Mbps and SCTP(0s-120s) is about 18Mbps. (The bandwidth(20Mbps) and delay(200ms) is set by dummynet) The SCTP and TCP are tested in same environment. ​ sctp.pca

Re: A problem on TCP in High RTT Environment.

Niu Zhixiong wrote this message on Sun, Aug 10, 2014 at 10:12 +0800: > During the TCP4 transmission. > Proto Recv-Q Send-Q Local Address Foreign Address(state) > tcp4 0 2097346 10.0.10.2.1350410.0.10.3.9000 > ESTABLISHED Ok, so you are getting a full 2MB in there, an

Re: A problem on TCP in High RTT Environment.

Hi During the TCP4 transmission. Proto Recv-Q Send-Q Local Address Foreign Address(state) tcp4 0 2097346 10.0.10.2.1350410.0.10.3.9000 ESTABLISHED Regards, Niu Zhixiong －－－ kaia...@gmail.com On Sun, Aug 10, 2014 at 4:58 AM, Michael Tuexen < michael.t

Re: A problem on TCP in High RTT Environment.

On 09 Aug 2014, at 22:45, John-Mark Gurney wrote: > Michael Tuexen wrote this message on Sat, Aug 09, 2014 at 21:51 +0200: >> >> On 09 Aug 2014, at 20:42, John-Mark Gurney wrote: >> >>> Niu Zhixiong wrote this message on Fri, Aug 08, 2014 at 20:34 +0800: Dear all, Last month,

Re: A problem on TCP in High RTT Environment.

On 09 Aug 2014, at 22:22, hiren panchasara wrote: > On Sat, Aug 9, 2014 at 12:51 PM, Michael Tuexen > wrote: >> >> On 09 Aug 2014, at 20:42, John-Mark Gurney wrote: >> >>> Niu Zhixiong wrote this message on Fri, Aug 08, 2014 at 20:34 +0800: Dear all, Last month, I send problem

Re: A problem on TCP in High RTT Environment.

Michael Tuexen wrote this message on Sat, Aug 09, 2014 at 21:51 +0200: > > On 09 Aug 2014, at 20:42, John-Mark Gurney wrote: > > > Niu Zhixiong wrote this message on Fri, Aug 08, 2014 at 20:34 +0800: > >> Dear all, > >> > >> Last month, I send problems related to FTP/TCP in a high RTT environme

Re: A problem on TCP in High RTT Environment.

On Sat, Aug 9, 2014 at 12:51 PM, Michael Tuexen wrote: > > On 09 Aug 2014, at 20:42, John-Mark Gurney wrote: > >> Niu Zhixiong wrote this message on Fri, Aug 08, 2014 at 20:34 +0800: >>> Dear all, >>> >>> Last month, I send problems related to FTP/TCP in a high RTT environment. >>> After that, I

Re: A problem on TCP in High RTT Environment.

On 09 Aug 2014, at 20:42, John-Mark Gurney wrote: > Niu Zhixiong wrote this message on Fri, Aug 08, 2014 at 20:34 +0800: >> Dear all, >> >> Last month, I send problems related to FTP/TCP in a high RTT environment. >> After that, I setup a simulation environment(Dummynet) to test TCP and SCTP >>

Re: A problem on TCP in High RTT Environment.

On 09 Aug 2014, at 11:51, Eggert, Lars wrote: > At 400ms at @ 20Mbps, your are probably receive window limited. Bump > net.inet.tcp.recvspace. (Your net.inet.sctp.recvspace is much larger, which > probably explains the performance difference.) The program he is using is netperfmeter and I think

Re: A problem on TCP in High RTT Environment.

Niu Zhixiong wrote this message on Fri, Aug 08, 2014 at 20:34 +0800: > Dear all, > > Last month, I send problems related to FTP/TCP in a high RTT environment. > After that, I setup a simulation environment(Dummynet) to test TCP and SCTP > in high delay environment. After finishing the test, I can

Re: A problem on TCP in High RTT Environment.

At 400ms at @ 20Mbps, your are probably receive window limited. Bump net.inet.tcp.recvspace. (Your net.inet.sctp.recvspace is much larger, which probably explains the performance difference.) On 2014-8-8, at 14:34, Niu Zhixiong wrote: > Dear all, > > Last month, I send problems related to FTP

A problem on TCP in High RTT Environment.

Dear all, Last month, I send problems related to FTP/TCP in a high RTT environment. After that, I setup a simulation environment(Dummynet) to test TCP and SCTP in high delay environment. After finishing the test, I can see TCP is always slower than SCTP. But, I think it is not possible. (Plz see t