| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-Id: <201503221614.47994.wrosner@tirnet.de> Date: Sun, 22 Mar 2015 16:14:46 +0100 From: Wolfgang Rosner <wrosner@...net.de> To: Eric Dumazet <eric.dumazet@...il.com> Cc: netdev@...r.kernel.org Subject: Re: One way TCP bottleneck over 6 x Gbit tecl aggregated link Hello, Eric, Am Sonntag, 22. März 2015 11:29:19 schrieben Sie: > On Sun, 2015-03-22 at 09:12 +0100, Wolfgang Rosner wrote: > > Am Samstag, 21. März 2015 22:48:22 schrieben Sie: > > > On Sat, 2015-03-21 at 12:32 +0100, Wolfgang Rosner wrote: > > > > btw: as it looks to me, the current git version seems to ignore the > > > > dsc filter. Sorry, you have to dig for the right line. > > > > > > Note I mentioned dst not dsc ;) > > > > sorry for the typo, but it's only in the mail text. > > In the script it says "dst" > > and I checked it again. > > > > Anyway, I think this tiny bug is not going to hurt us here, is it? > > > > Wolfgang > > > > > > =================================== > > > > root@...ncher:/home/tmp/netdev# > > /home/build/iproute2/shemminger/iproute2/misc/ss -temoi dst > > 192.168.130.225 > > State Recv-Q Send-Q Local > > Address:Port Peer Address:Port > > ESTAB 0 0 > > 192.168.0.53:ssh 192.168.0.50:21760 > > (....) > > ss -temoi dst 192.168.130.225 > > should not output tcp sessions to other destination than 192.168.130.225 > > Your example shows sessions to 192.168.0.50, this makes no sense to me. Agree. This is why I considered it as a bug. But just a sideline, not my main problem. And maybe it is even related to my build environment. Or to the fact that I call it with full path. or whatever.... I don't know whether this repo you pointed me https://git.kernel.org/cgit/linux/kernel/git/shemminger/iproute2.git/ is a bleeding edge developer fork, that is expected to show such fluctuations, or a should-be-stable mainstream one, where such behaviur is not supposed to occur. So if for you, there is reason to worry on this issue, we can try to debug into it. Let me know, and give me instructions for tests, if required. However, from my perspective, I'd prefer to return to my real problem: "One way TCP bottleneck over 6 x Gbit tecl aggregated link" If you got lost in the flood of netdev list (I just subscribed, and am really overwhelmed...), I don't mind to send you the info again and even remove the superfluous lines manually for you - see below. Do these figure tell you a story? Any hint where to look for a bottleneck? Presumaly related to some per-TCP-connection ressource? Wolfgang Rosner ########################################################################## # cat test.sh SSPATH=/home/build/iproute2/shemminger/iproute2/misc/ss SEP============================================================= (sleep 5 ; echo $SEP; $SSPATH -temoi dst 192.168.130.225 ; echo $SEP ) & nstat >/dev/null iperf -c 192.168.130.225 echo $SEP nstat # end of test script ######################################################################### # output follows, superflous lines from ss bug removed manually ------------------------------------------------------------ Client connecting to 192.168.130.225, TCP port 5001 TCP window size: 488 KByte (default) ------------------------------------------------------------ [ 3] local 192.168.130.250 port 39256 connected with 192.168.130.225 port 5001 ============================================================ # ....ss output - buggy lines removed .... State Recv-Q Send-Q Local Address:Port Peer Address:Port ESTAB 0 4554532 192.168.130.250:39256 192.168.130.225:5001 timer:(on,200ms,0) ino:514423 sk:ffff8801bdddd140 <-> skmem:(r0,rb500000,t66560,tb12000000,f3556096,w8469760,o0,bl3072) ts sack cubic wscale:7,10 rto:204 rtt:0.357/0.01 mss:8948 cwnd:956 ssthresh:17 send 191692.7Mbps lastsnd:4 lastrcv:140569508 lastack:4 unacked:16 retrans:0/1 sacked:1 reordering:22 rcv_space:26880 # .... buggy lines removed .... ============================================================ [ ID] Interval Transfer Bandwidth [ 3] 0.0-10.0 sec 3.78 GBytes 3.24 Gbits/sec ============================================================ #kernel IpInReceives 444749 0.0 IpForwDatagrams 4 0.0 IpInDelivers 444745 0.0 IpOutRequests 453019 0.0 TcpActiveOpens 1 0.0 TcpInSegs 444742 0.0 TcpOutSegs 453013 0.0 TcpRetransSegs 1 0.0 UdpInDatagrams 1 0.0 UdpOutDatagrams 1 0.0 UdpIgnoredMulti 2 0.0 TcpExtTCPHPHits 12 0.0 TcpExtTCPPureAcks 125155 0.0 TcpExtTCPHPAcks 71412 0.0 TcpExtTCPSackRecovery 1 0.0 TcpExtTCPSACKReorder 9 0.0 TcpExtTCPTSReorder 1 0.0 TcpExtTCPPartialUndo 1 0.0 TcpExtTCPFastRetrans 1 0.0 TcpExtTCPDSACKRecv 1 0.0 TcpExtTCPDSACKIgnoredNoUndo 1 0.0 TcpExtTCPSackShiftFallback 50003 0.0 TcpExtTCPAutoCorking 23882 0.0 TcpExtTCPOrigDataSent 453010 0.0 TcpExtTCPHystartTrainDetect 1 0.0 TcpExtTCPHystartTrainCwnd 17 0.0 IpExtInBcastPkts 3 0.0 IpExtOutBcastPkts 1 0.0 IpExtInOctets 24508230 0.0 IpExtOutOctets 4076970728 0.0 IpExtInBcastOctets 538 0.0 IpExtOutBcastOctets 76 0.0 IpExtInNoECTPkts 444749 0.0 -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majordomo@...r.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists