Network loss

[Rick Macklem]

Markus Gebert wrote:
> 
> On 28.02.2014, at 00:13, Rick Macklem <rmacklem at uoguelph.ca> wrote:
> 
> > Markus Gebert wrote:
> >> 
> >> On 27.02.2014, at 02:00, Rick Macklem <rmacklem at uoguelph.ca>
> >> wrote:
> >> 
> >>> John Baldwin wrote:
> >>>> On Tuesday, February 25, 2014 2:19:01 am Johan Kooijman wrote:
> >>>>> Hi all,
> >>>>> 
> >>>>> I have a weird situation here where I can't get my head around.
> >>>>> 
> >>>>> One FreeBSD 9.2-STABLE ZFS/NFS box, multiple Linux clients.
> >>>>> Once
> >>>>> in
> >>>>> a while
> >>>>> the Linux clients loose their NFS connection:
> >>>>> 
> >>>>> Feb 25 06:24:09 hv3 kernel: nfs: server 10.0.24.1 not
> >>>>> responding,
> >>>>> timed out
> >>>>> 
> >>>>> Not all boxes, just one out of the cluster. The weird part is
> >>>>> that
> >>>>> when I
> >>>>> try to ping a Linux client from the FreeBSD box, I have between
> >>>>> 10
> >>>>> and 30%
> >>>>> packetloss - all day long, no specific timeframe. If I ping the
> >>>>> Linux
> >>>>> clients - no loss. If I ping back from the Linux clients to
> >>>>> FBSD
> >>>>> box - no
> >>>>> loss.
> >>>>> 
> >>>>> The errors I get when pinging a Linux client is this one:
> >>>>> ping: sendto: File too large
> >> 
> >> We were facing similar problems when upgrading to 9.2 and have
> >> stayed
> >> with 9.1 on affected systems for now. We’ve seen this on HP G8
> >> blades with 82599EB controllers:
> >> 
> >> ix0 at pci0:4:0:0:	class=0x020000 card=0x18d0103c chip=0x10f88086
> >> rev=0x01 hdr=0x00
> >>    vendor     = 'Intel Corporation'
> >>    device     = '82599EB 10 Gigabit Dual Port Backplane
> >>    Connection'
> >>    class      = network
> >>    subclass   = ethernet
> >> 
> >> We didn’t find a way to trigger the problem reliably. But when it
> >> occurs, it usually affects only one interface. Symptoms include:
> >> 
> >> - socket functions return the 'File too large' error mentioned by
> >> Johan
> >> - socket functions return 'No buffer space’ available
> >> - heavy to full packet loss on the affected interface
> >> - “stuck” TCP connection, i.e. ESTABLISHED TCP connections that
> >> should have timed out stick around forever (socket on the other
> >> side
> >> could have been closed ours ago)
> >> - userland programs using the corresponding sockets usually got
> >> stuck
> >> too (can’t find kernel traces right now, but always in network
> >> related syscalls)
> >> 
> >> Network is only lightly loaded on the affected systems (usually
> >> 5-20
> >> mbit, capped at 200 mbit, per server), and netstat never showed
> >> any
> >> indication of ressource shortage (like mbufs).
> >> 
> >> What made the problem go away temporariliy was to ifconfig down/up
> >> the affected interface.
> >> 
> >> We tested a 9.2 kernel with the 9.1 ixgbe driver, which was not
> >> really stable. Also, we tested a few revisions between 9.1 and 9.2
> >> to find out when the problem started. Unfortunately, the ixgbe
> >> driver turned out to be mostly unstable on our systems between
> >> these
> >> releases, worse than on 9.2. The instability was introduced
> >> shortly
> >> after to 9.1 and fixed only very shortly before 9.2 release. So no
> >> luck there. We ended up using 9.1 with backports of 9.2 features
> >> we
> >> really need.
> >> 
> >> What we can’t tell is wether it’s the 9.2 kernel or the 9.2 ixgbe
> >> driver or a combination of both that causes these problems.
> >> Unfortunately we ran out of time (and ideas).
> >> 
> >> 
> >>>> EFBIG is sometimes used for drivers when a packet takes too many
> >>>> scatter/gather entries.  Since you mentioned NFS, one thing you
> >>>> can
> >>>> try is to
> >>>> disable TSO on the intertface you are using for NFS to see if
> >>>> that
> >>>> "fixes" it.
> >>>> 
> >>> And please email if you try it and let us know if it helps.
> >>> 
> >>> I've think I've figured out how 64K NFS read replies can do this,
> >>> but I'll admit "ping" is a mystery? (Doesn't it just send a
> >>> single
> >>> packet that would be in a single mbuf?)
> >>> 
> >>> I think the EFBIG is replied by bus_dmamap_load_mbuf_sg(), but I
> >>> don't know if it can happen for an mbuf chain with < 32 entries?
> >> 
> >> We don’t use the nfs server on our systems, but they’re
> >> (new)nfsclients. So I don’t think our problem is nfs related,
> >> unless
> >> the default rsize/wsize for client mounts is not 8K, which I
> >> thought
> >> it was. Can you confirm this, Rick?
> >> 
> > Well, if you don't specify any mount options, it will be
> > min(64K, what-the-server-specifies).
> > 
> > "nfsstat -m" should show you what it actually is using, for 9.2 or
> > later.
> 
> Thanks for your answer and the command. I knew there is an new option
> to nfsstat, but I couldn’t find it in the 9.1 man page, of course
> ;-).
> 
> 
> > 8K would be used if you specified "udp”.
> 
> I see. We’re using tcp, so that would not be relevant then. Guess my
> look at the mount vfsop was too quick...
> 
> 
> > For the client, it would be write requests that could be 64K.
> > You could try "wsize=32768,rsize=32768" (it is actually the
> > wsize that matters for this case, but you might as well set
> > rsize at the same time). With these options specified, you
> > know what the maximum value is (it will still be reduced for
> > udp or if the server wants it smaller).
> 
> Ok. I checked this on a system that is still running under 9.2 (we
> had to downgrade most production systems), and bingo, 64K wsize. So
> our nfs server (a NetApp Cluster) seems to prefer 64K.
> 
I suspect (don't know) that the NetApp Cluster prefers larger than 64K,
but MAXBSIZE is set to 64K on FreeBSD at this time. (I do know that
Solaris10 prefers 256K and handles 1M for UFS exports.)
Sorry, this is basically irrelevant to this discussion, but I just
wanted to note to readers that NFS servers like big IO sizes.
(Or at least they say they do;-)

> This means NFS still is a potential trigger of the problem. Let’s
> pretend it is, despite I think I already tried disabling TSO. Would
> this explain all the symptoms we were seeing? Why would ifconfig
> down/up help? Do you have a theory on that?
> 
I have no idea what effect ifconfig down/up would have. It certainly
shouldn't be relevant to the nfs related issue.

Btw, if you are not running out of mbuf clusters, all the nfs issue
does is introduce overhead from the m_defrag() calls. It is only if
m_defrag() fails to copy the write request to 32 mbufs that things
should fail (from my quick reading of the driver). As such, I have no
idea if it is a trigger or not?

rick

> 
> Markus
> 
> 
> _______________________________________________
> 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"
>