Date: Thu, 08 Oct 2009 17:14:01 +0500 From: rihad <rihad@mail.ru> To: Robert Watson <rwatson@FreeBSD.org> Cc: freebsd-net@freebsd.org, Eugene Grosbein <eugen@kuzbass.ru>, Luigi Rizzo <rizzo@iet.unipi.it>, Julian Elischer <julian@elischer.org> Subject: Re: dummynet dropping too many packets Message-ID: <4ACDD789.8030804@mail.ru> In-Reply-To: <alpine.BSF.2.00.0910071420290.38033@fledge.watson.org> References: <4AC9E29B.6080908@mail.ru> <20091005123230.GA64167@onelab2.iet.unipi.it> <4AC9EFDF.4080302@mail.ru> <4ACA2CC6.70201@elischer.org> <4ACAFF2A.1000206@mail.ru> <4ACB0C22.4000008@mail.ru> <20091006100726.GA26426@svzserv.kemerovo.su> <4ACB42D2.2070909@mail.ru> <20091006142152.GA42350@svzserv.kemerovo.su> <4ACB6223.1000709@mail.ru> <20091006161240.GA49940@svzserv.kemerovo.su> <alpine.BSF.2.00.0910061804340.50283@fledge.watson.org> <4ACC5563.602@mail.ru> <4ACC56A6.1030808@mail.ru> <alpine.BSF.2.00.0910070957430.58146@fledge.watson.org> <4ACC5DEC.1010006@mail.ru> <alpine.BSF.2.00.0910071036280.58146@fledge.watson.org> <4ACC65A0.7030900@mail.ru> <alpine.BSF.2.00.0910071312420.58146@fledge.watson.org> <4ACC8CC8.8050403@mail.ru> <alpine.BSF.2.00.0910071420290.38033@fledge.watson.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Robert Watson wrote:
> I would suggest making just the HZ -> 4000 change for now and see how it
> goes.
>
Been running for a few hours under these changed sysctls:
kern.clockrate: { hz = 4000, tick = 250, profhz = 4000, stathz = 129 }
net.inet.ip.dummynet.io_fast: 1
net.inet.ip.dummynet.hash_size: 512
net.isr.direct: 0
net.inet.ip.intr_queue_maxlen: 5000
kern.ipc.nmbclusters=111111
Current stats:
net.inet.ip.dummynet.search_steps: 190347891
net.inet.ip.dummynet.searches: 188979871
net.inet.ip.dummynet.io_pkt_drop: 0
net.inet.ip.intr_queue_drops: 0
Around 1800 entries in each of the two ipfw tables.
About 420 mbps of traffic passing through.
ipfw pipe's created with GRED queuing algorithm, if it matters:
ipfw pipe 64 config bw 64kbit/s mask dst-ip 0xffffffff gred
0.002/1800/2000/0.1 queue 2000
etc.
0 global output packet drops per second.
top -SH:
last pid: 59120; load averages: 0.87, 1.19, 1.27
up 0+04:31:38 17:11:59
141 processes: 9 running, 115 sleeping, 17 waiting
CPU: 3.8% user, 0.0% nice, 3.0% system, 8.8% interrupt, 84.4% idle
Mem: 1302M Active, 1482M Inact, 350M Wired, 40K Cache, 214M Buf, 801M Free
Swap: 2048M Total, 2048M Free
PID USERNAME PRI NICE SIZE RES STATE C TIME WCPU COMMAND
16 root 171 ki31 0K 16K CPU2 2 237:10 97.02% idle: cpu2
14 root 171 ki31 0K 16K CPU4 4 225:33 96.48% idle: cpu4
12 root 171 ki31 0K 16K CPU6 6 249:31 88.96% idle: cpu6
15 root 171 ki31 0K 16K CPU3 3 224:18 88.92% idle: cpu3
13 root 171 ki31 0K 16K CPU5 5 228:39 87.74% idle: cpu5
11 root 171 ki31 0K 16K CPU7 7 235:02 87.35% idle: cpu7
18 root 171 ki31 0K 16K RUN 0 198:11 81.98% idle: cpu0
17 root 171 ki31 0K 16K CPU1 1 209:31 71.78% idle: cpu1
21 root -44 - 0K 16K WAIT 0 88:24 47.31% swi1: net
29 root -68 - 0K 16K WAIT 1 34:06 18.95% irq256: bce0
470 root -68 - 0K 16K - 7 21:24 6.01% dummynet
19 root -32 - 0K 16K WAIT 5 13:37 5.86% swi4:
clock sio
31 root -68 - 0K 16K WAIT 2 7:45 4.44% irq257: bce1
I'll wait for the number of entries to reach 2000-2200 or so, and post
back again.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4ACDD789.8030804>