9.1-RELEASE-p1: em0: Could not setup receive structures

Fri Mar 8 20:20:09 UTC 2013

kern.ipc.nmbclusters

Jeff

-----Original Message-----
From: owner-freebsd-net at freebsd.org [mailto:owner-freebsd-net at freebsd.org] On Behalf Of mxb
Sent: Friday, March 08, 2013 12:17 PM
To: Jack Vogel
Cc: freebsd-net at freebsd.org; mxb
Subject: Re: 9.1-RELEASE-p1: em0: Could not setup receive structures

Any sysctl I'd should look out for?

//maxim

On 8 mar 2013, at 21:03, Jack Vogel <jfvogel at gmail.com> wrote:

> The message occurs because you don't have enough mbufs to setup the RX
> ring, so you
> need to look at nmbclusters. It may be that em is just the victim, since
> you have igb interfaces
> as well from what I see.
> 
> Jack
> 
> 
> On Fri, Mar 8, 2013 at 11:19 AM, mxb <mxb at alumni.chalmers.se> wrote:
> 
>> 
>> Hello list@,
>> 
>> I'm mostly active on OpenBSD-side, however I have several machines running
>> fbsd with ZFS.
>> 
>> I'v recently upgraded(today) from 8.2-stable to 9.1-rel because of em(4)
>> with the problem  <subject> on 8.2-stable.
>> However, my problem has not disappeared after mentioned upgrade.
>> 
>> I serve VMWare images from this machine.
>> 
>> Configuration:
>> 
>> lagg0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 9000
>> 
>> options=4019b<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,VLAN_HWCSUM,TSO4,VLAN_HWTSO>
>>        ether 00:25:90:24:70:e8
>>        inet 172.16.0.243 netmask 0xfffff800 broadcast 172.16.7.255
>>        nd6 options=1<PERFORMNUD>
>>        media: Ethernet autoselect
>>        status: active
>>        laggproto lacp lagghash l2,l3,l4
>>        laggport: igb0 flags=1c<ACTIVE,COLLECTING,DISTRIBUTING>
>>        laggport: em0 flags=18<COLLECTING,DISTRIBUTING>
>> lagg1: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 9000
>> 
>> options=4019b<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,VLAN_HWCSUM,TSO4,VLAN_HWTSO>
>>        ether 00:25:90:24:70:e9
>>        inet 10.11.11.1 netmask 0xffffff00 broadcast 10.11.11.255
>>        inet 10.11.11.11 netmask 0xffffff00 broadcast 10.11.11.255
>>        inet 10.11.11.12 netmask 0xffffff00 broadcast 10.11.11.255
>>        inet 10.11.11.13 netmask 0xffffff00 broadcast 10.11.11.255
>>        inet 10.11.11.14 netmask 0xffffff00 broadcast 10.11.11.255
>>        inet 10.11.11.15 netmask 0xffffff00 broadcast 10.11.11.255
>>        inet 10.11.11.16 netmask 0xffffff00 broadcast 10.11.11.255
>>        inet 10.11.11.17 netmask 0xffffff00 broadcast 10.11.11.255
>>        inet 10.11.11.18 netmask 0xffffff00 broadcast 10.11.11.255
>>        inet 10.11.11.19 netmask 0xffffff00 broadcast 10.11.11.255
>>        inet 10.11.11.20 netmask 0xffffff00 broadcast 10.11.11.255
>>        inet 10.11.11.21 netmask 0xffffff00 broadcast 10.11.11.255
>>        inet 10.11.11.22 netmask 0xffffff00 broadcast 10.11.11.255
>>        inet 10.11.11.23 netmask 0xffffff00 broadcast 10.11.11.255
>>        inet 10.11.11.24 netmask 0xffffff00 broadcast 10.11.11.255
>>        inet 10.11.11.25 netmask 0xffffff00 broadcast 10.11.11.255
>>        inet 10.11.11.26 netmask 0xffffff00 broadcast 10.11.11.255
>>        inet 10.11.11.27 netmask 0xffffff00 broadcast 10.11.11.255
>>        inet 10.11.11.28 netmask 0xffffff00 broadcast 10.11.11.255
>>        nd6 options=1<PERFORMNUD>
>>        media: Ethernet autoselect
>>        status: active
>>        laggproto lacp lagghash l2,l3,l4
>>        laggport: igb1 flags=1c<ACTIVE,COLLECTING,DISTRIBUTING>
>>        laggport: em1 flags=1c<ACTIVE,COLLECTING,DISTRIBUTING>
>> 
>> current sysctl.conf (not fixed after upgrade):
>> # Every socket is a file, so increase them
>> kern.maxfiles=204800
>> kern.maxfilesperproc=200000
>> kern.ipc.maxsockets=204800
>> 
>> # Increase max command-line length showed in `ps` (e.g for Tomcat/Java)
>> # Default is PAGE_SIZE / 16 or 256 on x86
>> # For more info see: http://www.freebsd.org/cgi/query-pr.cgi?pr=120749
>> kern.ps_arg_cache_limit=4096
>> 
>> # Security
>> #net.inet.udp.blackhole=1
>> #net.inet.tcp.blackhole=2
>> 
>> kern.ipc.maxsockbuf=16777216
>> kern.ipc.nmbclusters=65535
>> kern.ipc.somaxconn=32768
>> #kern.maxfiles=65535
>> kern.maxvnodes=800000
>> 
>> vfs.zfs.l2arc_noprefetch=0
>> vfs.zfs.l2arc_write_max=16777216
>> vfs.zfs.l2arc_write_boost=16777216
>> 
>> net.inet.tcp.sendspace=65535
>> net.inet.tcp.recvspace=131072
>> net.inet.tcp.mssdflt=1452
>> net.inet.tcp.sendbuf_max=16777216
>> net.inet.tcp.sendbuf_inc=524288
>> net.inet.tcp.recvbuf_max=16777216
>> net.inet.tcp.recvbuf_inc=524288
>> net.inet.udp.recvspace=65535
>> net.inet.udp.maxdgram=65535
>> net.local.stream.recvspace=65535
>> net.local.stream.sendspace=65535
>> net.inet.tcp.delayed_ack=0
>> 
>> 
>> Any clues?
>> 
>> Please mail directly to me or cc to sysop at prisjakt.nu
>> 
>> Regards
>> Maxim
>> 
>> 
>> _______________________________________________
>> freebsd-net at freebsd.org mailing list
>> http://lists.freebsd.org/mailman/listinfo/freebsd-net
>> To unsubscribe, send any mail to "freebsd-net-unsubscribe at freebsd.org"
>> 
> _______________________________________________
> freebsd-net at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-net
> To unsubscribe, send any mail to "freebsd-net-unsubscribe at freebsd.org"

_______________________________________________
freebsd-net at freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "freebsd-net-unsubscribe at freebsd.org"