standards/175711: When the server has more than 3 days, rising interrupt (serving Pppoe.) Traffic and pppoe session has not changed.
Artem Goncharenko
roysbike at gmail.com
Wed Jan 30 20:10:00 UTC 2013
>Number: 175711
>Category: standards
>Synopsis: When the server has more than 3 days, rising interrupt (serving Pppoe.) Traffic and pppoe session has not changed.
>Confidential: no
>Severity: critical
>Priority: medium
>Responsible: freebsd-standards
>State: open
>Quarter:
>Keywords:
>Date-Required:
>Class: sw-bug
>Submitter-Id: current-users
>Arrival-Date: Wed Jan 30 20:10:00 UTC 2013
>Closed-Date:
>Last-Modified:
>Originator: Artem Goncharenko
>Release: frebsd 9.1 stable
>Organization:
òÕÓÉÞô÷î
>Environment:
FreeBSD pppoe2.rutvn.ru 9.1-STABLE FreeBSD 9.1-STABLE #2 r245564: Mon Jan 21 00:21:44 UTC 2013 root at pppoe2.rutvn.ru:/usr/obj/usr/src/sys/GENERIC.ok amd64
>Description:
Good afternoon. I'm using freebsd + mpd5.6 for router. PPPoE client for 1500-2000. On the server, network cards are
igb0 at pci0:10:0:0: class=0x020000 card=0xa03c8086 chip=0x10c98086 rev=0x01 hdr=0x00
vendor = 'Intel Corporation'
device = '82576 Gigabit Network Connection'
class = network
subclass = ethernet
igb1 at pci0:10:0:1: class=0x020000 card=0xa03c8086 chip=0x10c98086 rev=0x01 hdr=0x00
vendor = 'Intel Corporation'
device = '82576 Gigabit Network Connection'
class = network
subclass = ethernet
bge0 at pci0:3:4:0: class=0x020000 card=0x703e103c chip=0x167814e4 rev=0xa3 hdr=0x00
vendor = 'Broadcom Corporation'
device = 'NetXtreme BCM5715 Gigabit Ethernet'
class = network
subclass = ethernet
bge1 at pci0:3:4:1: class=0x020000 card=0x703e103c chip=0x167814e4 rev=0xa3 hdr=0x00
vendor = 'Broadcom Corporation'
device = 'NetXtreme BCM5715 Gigabit Ethernet'
class = network
subclass = ethernet
igb0 + igb1 = lagg0, bge0 + bge1 = lagg1. On lagg1 100 vlan, hears all, mpd (PPPoE).
Queues nailed to each core. igb0 = cpu0 igb1 = cpu1 bge0 = cpu2 bge1 = cpu3. Next problem. When the server is live one day, the same interrupt.
Uptime one day
CPU 0: 3.2% user, 0.0% nice, 1.3% system, 25.0% interrupt, 70.5% idle
CPU 1: 1.3% user, 0.0% nice, 0.6% system, 30.8% interrupt, 67.3% idle
CPU 2: 1.3% user, 0.0% nice, 1% system, 35.5% interrupt, 62.2% idle
CPU 3: 1.9% user, 0.0% nice, 1% system, 40.0% interrupt, 57.1% idle
When uptime > 3 days interrupt bge0 and bge1 begin to grow. Traffic is not changed and pps too. 1500-1800 the number of clients
Server uptime 9 days
last pid: 7979; load averages: 1.05, 1.16, 1.29 up 9+13:35:15 23:42:36
152 processes: 5 running, 118 sleeping, 29 waiting
CPU 0: 4.9% user, 0.0% nice, 1.2% system, 14.8% interrupt, 79.0% idle
CPU 1: 1.2% user, 0.0% nice, 8.6% system, 7.4% interrupt, 82.7% idle
CPU 2: 1.2% user, 0.0% nice, 0.0% system, 49.4% interrupt, 49.4% idle
CPU 3: 0.0% user, 0.0% nice, 2.5% system, 38.3% interrupt, 59.3% idle
Mem: 62M Active, 462M Inact, 759M Wired, 622M Buf, 4645M Free
Swap: 4096M Total, 4096M Free
PID USERNAME PRI NICE SIZE RES STATE C TIME WCPU COMMAND
11 root 155 ki31 0K 64K CPU0 0 190.9H 94.38% idle{idle: cpu0}
11 root 155 ki31 0K 64K RUN 1 192.5H 91.16% idle{idle: cpu1}
11 root 155 ki31 0K 64K CPU3 3 140.1H 65.67% idle{idle: cpu3}
11 root 155 ki31 0K 64K CPU2 2 144.4H 56.49% idle{idle: cpu2}
12 root -92 - 0K 464K WAIT 2 82.5H 45.26% intr{irq266: bge0}
12 root -92 - 0K 464K WAIT 3 86.1H 33.89% intr{irq267: bge1}
12 root -92 - 0K 464K WAIT 0 533:30 4.30% intr{irq256: igb0:que}
12 root -92 - 0K 464K WAIT 0 459:38 2.20% intr{irq257: igb0:que}
12 root -92 - 0K 464K WAIT 1 494:31 1.86% intr{irq264: igb1:que}
12 root -92 - 0K 464K WAIT 0 503:23 1.76% intr{irq259: igb0:que}
12 root -92 - 0K 464K WAIT 0 493:21 1.46% intr{irq258: igb0:que}
12 root -92 - 0K 464K WAIT 1 484:58 1.37% intr{irq263: igb1:que}
10520 root 21 0 37596K 10164K select 0 493:54 1.27% snmpd
12 root -92 - 0K 464K WAIT 1 456:25 1.27% intr{irq262: igb1:que}
12 root -92 - 0K 464K WAIT 1 450:34 0.78% intr{irq261: igb1:que}
99079 root 22 2 128M 62820K select 1 17:52 0.29% mpd5{mpd5}
7978 nagios 25 0 20472K 1728K pause 1 0:00 0.00% netstat
pppoe2# procstat -at | grep bge
12 100069 intr irq266: bge0 2 8 wait -
12 100070 intr irq267: bge1 3 8 run -
pppoe2# procstat -at | grep ": igb"
12 100032 intr irq256: igb0:que 0 8 wait -
12 100034 intr irq257: igb0:que 0 8 wait -
12 100036 intr irq258: igb0:que 0 8 wait -
12 100038 intr irq259: igb0:que 0 8 wait -
12 100040 intr irq260: igb0:lin 1 8 wait -
12 100041 intr irq261: igb1:que 1 8 wait -
12 100043 intr irq262: igb1:que 1 8 wait -
12 100045 intr irq263: igb1:que 1 8 wait -
12 100047 intr irq264: igb1:que 1 8 wait -
12 100049 intr irq265: igb1:lin 3 8 wait -
pppoe2# ifconfig lagg0
lagg0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500
options=401bb<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,JUMBO_MTU,VLAN_HWCSUM,TSO4,VLAN_HWTSO>
ether 90:e2:ba:00:a7:7a
inet 172.30.0.2 netmask 0xffff0000 broadcast 172.30.255.255
media: Ethernet autoselect
status: active
laggproto lacp
laggport: igb1 flags=1c<ACTIVE,COLLECTING,DISTRIBUTING>
laggport: igb0 flags=1c<ACTIVE,COLLECTING,DISTRIBUTING>
lagg1: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500
options=8009b<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,VLAN_HWCSUM,LINKSTATE>
ether 10:1f:74:3b:eb:ec
media: Ethernet autoselect
status: active
laggproto lacp
laggport: bge1 flags=1c<ACTIVE,COLLECTING,DISTRIBUTING>
laggport: bge0 flags=1c<ACTIVE,COLLECTING,DISTRIBUTING>
lagg1 and 100 vlan's
MPD5.6 Listen 100 vlan for PPPoE.
What other parameters of the system to send?
>How-To-Repeat:
When the server has more than 3 days, begins to rise interrupt. After reboot interrupt again smooth and the server can produce more than 1100 Mbps at 2000 clients. Why can grow interrupt, if the traffic is not changed and the number of pppoe sessions too. In my problem only helps reboot.
>Fix:
>Release-Note:
>Audit-Trail:
>Unformatted:
More information about the freebsd-standards
mailing list