Hi,

previously, with SYNPROXY and NIC firmware version 16.0.21 we experienced 3
reboots in a row after 3 or 3½ days each. Now, with SYNPROXY and NIC
firmware version 16.5.20 the server has been running since 6½ days without
a random reboot. So the firmware update seems to have solved the problem!

I have nevertheless uploaded the current ethtool statistics:
http://pastebin.com/fz58F73t

I'm not sure which of the fixes in 16.5 could possibly have solved our
issue:
http://www.dell.com/support/home/us/en/04/Drivers/DriversDetails?driverId=6FD9P

The other option is that our kernel upgrade from 4.1.13 to 4.1.15 has fixed
the problem. There do not seem to be any SYNPROXY or Netfilter fixes in
4.1.14 and 4.1.15 but I'm not sure whether any of the other patches in
4.1.14 or 4.1.15 could possibly have solved this issue.

Thanks,
Steffen


On Tue, Dec 8, 2015 at 10:09 PM, Skidmore, Donald C <
donald.c.skidm...@intel.com> wrote:

> This brings up an interesting.  It would be nice to see the ethtool stats
> right before the reset.  It might be interesting to grab those stats on a
> system that has been running for a few days, that you "expect" will fail in
> the future.
>
> > -----Original Message-----
> > From: Brandeburg, Jesse
> > Sent: Tuesday, December 08, 2015 12:36 PM
> > To: Steffen Weber; Skidmore, Donald C; e1000-devel@lists.sourceforge.net
> > Subject: RE: [E1000-devel] ixgbe and using iptables/SYNPROXY causes
> > random system resets
> >
> > Also because the problem is maybe related to SYN, maybe trying turning
> off
> > flow-director/ATR by doing something like:
> >
> > ethtool -K ethx ntuple on
> >
> >
> > -----Original Message-----
> > From: Steffen Weber [mailto:steffen.we...@gmail.com]
> > Sent: Tuesday, December 08, 2015 12:12 PM
> > To: Skidmore, Donald C <donald.c.skidm...@intel.com>; e1000-
> > de...@lists.sourceforge.net
> > Subject: Re: [E1000-devel] ixgbe and using iptables/SYNPROXY causes
> > random system resets
> >
> > Hi Don,
> >
> > thank you for the reply.
> >
> > We've been using the X520 NICs since February 2015 and so far they have
> > been working fine. A few weeks ago we enabled SYNPROXY and the reboots
> > started to happen. So, only judging from our own experience, this looks
> > more like a SYNPROXY problem than an X520/ixgbe problem. But the author
> > of the thread I'm replying to has described exactly the same issue that
> we are
> > seeing with the combination of X520 NICs and SYNPROXY (sudden
> > reboots) and he was only able to reproduce the issue with X520 NICs but
> not
> > with other NICs. That's the (only) reason why I'm assuming that the
> > X520 hardware/firmware or the ixgbe driver may have something to with the
> > problem.
> >
> > Our system BIOS is up-to-date (version 2.4.2), as is the Lifecycle
> Controller
> > firmware (version 2.21.21.21). We are not using any out-of-tree drivers,
> just
> > pure Linux 4.1.13.
> >
> > http://www.dell.com/support/home/us/en/04/product-
> > support/product/poweredge-r420/drivers
> >
> > Is the "latest SourceForge driver" part of Linux 4.2 or 4.3? I have more
> > routine in switching kernels than in installing out-of-tree drivers.
> >
> > I agree that just rebooting is a rather strange symptom of a bug. I'll
> let you
> > know whether upgrading the firmware and/or disabling GRO/LRO helps (in
> > the coming days / weeks).
> >
> > Thanks,
> > Steffen
> >
> > On 08.12.2015 19:53, Skidmore, Donald C wrote:
> > > Hey Steffen,
> > >
> > > Sorry to hear about your system reset problem.  Since you're not seeing
> > anything in the system logs is you thinking that the NIC's are involved
> due to
> > the frequency of the failures seems to correspond to amount of system
> > traffic?  I'm asking as in my 7 years of maintaining the ixgbe driver
> this failure
> > state is new to me.  You could try loading the latest Source Forge driver
> > verify this still fails with a slightly more recent driver, but really
> there
> > shouldn't be much in the way for differences and nothing I would expect
> to
> > correct this sort of failure.  It almost sounds like a power issue, the
> way the
> > system resets without even logging the problem, is your system BOIS up to
> > date?
> > >
> > > I know this isn't much in the way of help, but you have me at a bit of
> a loss
> > on how the NIC could lead to this sort of immediate system reset.
> > >
> > > Thanks,
> > > -Don Skidmore <donald.c.skidm...@intel.com>
> > >
> > >> -----Original Message-----
> > >> From: Steffen Weber [mailto:steffen.we...@gmail.com]
> > >> Sent: Tuesday, December 08, 2015 5:03 AM
> > >> To: e1000-devel@lists.sourceforge.net
> > >> Subject: Re: [E1000-devel] ixgbe and using iptables/SYNPROXY causes
> > >> random system resets
> > >>
> > >> Hi,
> > >>
> > >> we have exactly the same problem with the combination of SYNPROXY
> > and
> > >> X520 NICs. We use Dell PowerEdge R420 servers.
> > >>
> > >> CPU: Intel(R) Xeon(R) CPU E5-2430
> > >> NIC: Intel Corporation Ethernet 10G 2P X520 Adapter (rev 01)
> > >> Kernel: Linux 4.1.13 with in-tree ixgbe 4.0.1-k
> > >>
> > >> One of those servers with relatively constant TCP traffic on ports 80
> > >> +
> > >> 443 has rebooted three times, each time after about 3 days of uptime.
> > >> Another server with less constant traffic has rebooted after 7 days.
> > >> Two other servers with way less network traffic did not reboot within
> 7
> > days.
> > >>
> > >> There is nothing in the logs when the reboots happen.
> > >>
> > >> We've just updated the NIC firmware from version 16.0.21 to version
> > >> 16.5.20 but according to the changelog supplied by Dell this probably
> > >> won't fix the problem. (I'll let you know in a few days.)
> > >>
> > >>
> > http://www.dell.com/support/home/us/en/19/Drivers/DriversDetails?driv
> > >> erI
> > >> d=6FD9P
> > >>
> > >> Is there anything we can do?
> > >>
> > >> Thanks,
> > >> Steffen
> >
> >
> ------------------------------------------------------------------------------
> > Go from Idea to Many App Stores Faster with Intel(R) XDK Give your users
> > amazing mobile app experiences with Intel(R) XDK.
> > Use one codebase in this all-in-one HTML5 development environment.
> > Design, debug & build mobile apps & 2D/3D high-impact games for multiple
> > OSs.
> > http://pubads.g.doubleclick.net/gampad/clk?id=254741911&iu=/4140
> > _______________________________________________
> > E1000-devel mailing list
> > E1000-devel@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/e1000-devel
> > To learn more about Intel&#174; Ethernet, visit
> > http://communities.intel.com/community/wired
>
------------------------------------------------------------------------------
_______________________________________________
E1000-devel mailing list
E1000-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/e1000-devel
To learn more about Intel&#174; Ethernet, visit 
http://communities.intel.com/community/wired

Reply via email to