Date: Wed, 5 Aug 2009 14:42:05 -0700 (PDT) From: alexpalias-bsdstable@yahoo.com To: freebsd-stable@freebsd.org, "Marat N.Afanasyev" <amarat@ksu.ru> Subject: Re: em driver input errors Message-ID: <121465.1568.qm@web56402.mail.re3.yahoo.com> In-Reply-To: <4A79EFE6.5080001@ksu.ru>
next in thread | previous in thread | raw e-mail | index | archive | help
--- On Wed, 8/5/09, Marat N.Afanasyev <amarat@ksu.ru> wrote:
> From: Marat N.Afanasyev <amarat@ksu.ru>
> Subject: Re: em driver input errors
> To: alexpalias-bsdstable@yahoo.com, "freebsd-stable@freebsd.org >> FreeBSD-STABLE Mailing List" <freebsd-stable@freebsd.org>
> Date: Wednesday, August 5, 2009, 11:47 PM
> alexpalias-bsdstable@yahoo.com
> wrote:
> >
> > kern.ipc.nmbclusters=655360
> >
> do you have enough mbufs? look for mbuf clusters value in
> netstat -m output. if 'current' is near to 'total' you'll
> have drops of input packets. and take a look to 'jumbo
> clusters' in netstat -m too.
>
> -- SY, Marat
>
As mentioned in my initial mail, I did tune nmbclusters (you might say it's excessively high, 655360, but I still have 2.5 GB free memory on that machine. I forgot to mention, MTU is 1500 on all vlans, so there's no need for jumbo frames, only VLAN_MTU flag on the interfaces. All interfaces have:
em0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500
options=9b<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,VLAN_HWCSUM>
ether 00:14:22:xx:xx:xx
media: Ethernet autoselect (1000baseTX <full-duplex>)
status: active
Here's the output from netstat -m for reference:
> netstat -m
16390/9350/25740 mbufs in use (current/cache/total)
16388/7938/24326/655360 mbuf clusters in use (current/cache/total/max)
16388/7932 mbuf+clusters out of packet secondary zone in use (current/cache)
0/151/151/12800 4k (page size) jumbo clusters in use (current/cache/total/max)
0/0/0/6400 9k jumbo clusters in use (current/cache/total/max)
0/0/0/3200 16k jumbo clusters in use (current/cache/total/max)
36873K/18817K/55691K bytes allocated to network (current/cache/total)
0/0/0 requests for mbufs denied (mbufs/clusters/mbuf+clusters)
0/0/0 requests for jumbo clusters denied (4k/9k/16k)
0/0/0 sfbufs in use (current/peak/max)
0 requests for sfbufs denied
0 requests for sfbufs delayed
0 requests for I/O initiated by sendfile
0 calls to protocol drain routines
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?121465.1568.qm>