Date: Thu, 22 Dec 2011 19:34:18 -0500 From: Mike Tancsa <mike@sentex.net> To: =?windows-1251?Q?=CA=EE=ED=FC=EA=EE=E2_=C5=E2=E3=E5=ED=E8=E9?= <kes-kes@yandex.ru> Cc: freebsd-questions@freebsd.org Subject: Re: about igb queue Message-ID: <4EF3CC8A.8060103@sentex.net> In-Reply-To: <613871684.20111223000329@yandex.ru> References: <1444579738.20111222202238@yandex.ru> <613871684.20111223000329@yandex.ru>
next in thread | previous in thread | raw e-mail | index | archive | help
On 12/22/2011 5:03 PM, Êîíüêîâ Åâãåíèé wrote:
> Çäðàâñòâóéòå, Êîíüêîâ.
>
> Âû ïèñàëè 22 äåêàáðÿ 2011 ã., 20:22:38:
>
> ÊÅ> I have configured that I receive traffic from LAN via igb1 and it is
> ÊÅ> leaving to world via igb0
>
> ÊÅ> but when I see processes I see that on igb0 there is 4 queues but on
> ÊÅ> igb1 it is only one. Do I must tune something or all is right here?
vmstat -i
shows interrupts assigned to the nic. Whats the netgraph stuff doing ?
Perhaps provide more details about what the box is doing and do you have
any modules loaded. Do you have any tunables set in /etc/sysctl.conf or
in /boot/loader.conf ?
---Mike
>
> ÊÅ> top -SIHP
>
> ÊÅ> PID USERNAME PRI NICE SIZE RES STATE C TIME WCPU COMMAND
> ÊÅ> 11 root 155 ki31 0K 32K RUN 1 287:27 70.21% idle{idle: cpu1}
> ÊÅ> 11 root 155 ki31 0K 32K RUN 3 292:42 67.77% idle{idle: cpu3}
> ÊÅ> 11 root 155 ki31 0K 32K CPU2 2 286:38 65.97% idle{idle: cpu2}
> ÊÅ> 11 root 155 ki31 0K 32K RUN 0 282:58 57.13% idle{idle: cpu0}
> ÊÅ> 12 root -92 - 0K 248K WAIT 1 7:01 5.32% intr{irq257: igb0:que}
> ÊÅ> 12 root -92 - 0K 248K WAIT 0 9:35 5.03% intr{irq256: igb0:que}
> ÊÅ> 12 root -92 - 0K 248K WAIT 2 8:14 4.25% intr{irq258: igb0:que}
> ÊÅ> 12 root -92 - 0K 248K WAIT 3 9:48 3.71% intr{irq259: igb0:que}
> ÊÅ> 13 root -16 - 0K 32K sleep 2 6:42 3.08% ng_queue{ng_queue3}
> ÊÅ> 13 root -16 - 0K 32K sleep 3 6:42 2.98% ng_queue{ng_queue1}
> ÊÅ> 13 root -16 - 0K 32K sleep 0 6:42 2.93% ng_queue{ng_queue2}
> ÊÅ> 13 root -16 - 0K 32K sleep 2 6:43 2.69% ng_queue{ng_queue0}
> ÊÅ> 7371 root 21 0 15388K 5496K select 2 5:04 0.73% snmpd
> ÊÅ> 12 root -92 - 0K 248K WAIT 0 1:52 0.63% intr{irq261: igb1:que}
>
>
> getting information about interrupts shows that there is not
> interrupts from igb1
> char igb_driver_version[] = "version - 2.2.5";
>
>
> 2 users Load 0.50 0.53 0.50 Dec 22 23:59
>
> Mem:KB REAL VIRTUAL VN PAGER SWAP PAGER
> Tot Share Tot Share Free in out in out
> Act 332272 13268 2062288 30088 122004 count
> All 494672 19520 4333592 90716 pages
> Proc: Interrupts
> r p d s w Csw Trp Sys Int Sof Flt 117 cow 31201 total
> 1 172 72k 889 4135 14k 4041 660 296 zfod ata0 14
> 10 ozfod 5 ata1 15
> 3.9%Sys 5.0%Intr 0.4%User 0.0%Nice 90.7%Idle 3%ozfod 2 ehci0 16
> | | | | | | | | | | | daefr 2 ehci1 23
> ==++> 378 prcfr 4126 cpu0:timer
> 33 dtbuf 450 totfr 2920 igb0:que 0
> Namei Name-cache Dir-cache 110737 desvn react 2680 igb0:que 1
> Calls hits % hits % 37963 numvn pdwak 2187 igb0:que 2
> 4504 4492 100 27657 frevn pdpgs 2550 igb0:que 3
> intrn igb0:link
> Disks ada0 da0 pass0 pass1 270288 wire 4355 igb1:que 0
> KB/t 30.96 0.00 0.00 0.00 316488 act 3 igb1:que 1
> tps 5 0 0 0 1264188 inact 1 igb1:que 2
> MB/s 0.16 0.00 0.00 0.00 95204 cache 2 igb1:que 3
> %busy 0 0 0 0 26800 free igb1:link
> 114912 re0 266
> 4127 cpu1:timer
> 4116 cpu3:timer
> 4125 cpu2:timer
>
>
>
--
-------------------
Mike Tancsa, tel +1 519 651 3400
Sentex Communications, mike@sentex.net
Providing Internet services since 1994 www.sentex.net
Cambridge, Ontario Canada http://www.tancsa.com/
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4EF3CC8A.8060103>