lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20070506202234.GB17957@robsims.com>
Date:	Sun, 6 May 2007 14:22:34 -0600
From:	Rob Sims <lkml-z@...sims.com>
To:	Stephen Hemminger <shemminger@...ux-foundation.org>
Cc:	Thomas Glanzmann <thomas@...nzmann.de>,
	LKML <linux-kernel@...r.kernel.org>,
	Andreas Allgöwer <andreas.allgoewer@....de>,
	Michael Gernoth <simigern@....informatik.uni-erlangen.de>
Subject: Re: sky2 PHY setup

On Sat, Apr 07, 2007 at 11:53:48AM -0600, Rob Sims wrote:
> On Wed, Apr 04, 2007 at 11:19:30AM -0700, Stephen Hemminger wrote:
... 
> > You might have over run the hub and it wedged.  Try doing:
> > 	ethtool -r eth0
> > that forces a down/up
> 
> I'm still seeing throughput drop to under 1 Mb/s (50-100kB/s)
> periodically.  I did an ethtool -S and ethtool -d to capture state.
> (sky2-fail.log).  I ran ethtool -r and retested; no change
> (sky2-ethtool-r.log).  Finally, ifdown - rmmod - modprobe - ifup, which
> restored to Gigabit speeds (96+ MB/s), (sky2-modcycle.log).
> 
> Please let me know if there's anything else I can poke into.  As a
> reminder, this is a 88E8053 r20.  Next time I see a degradation, I'll
> try cycling the switch.

I continued to have problems with this interface.  I tried bringing up
the second interface on the motherboard; I added a cable to the same hub
as the first interface and configured an address on the same subnet.
I intended to see if I could use it as a redundant interface for when
the first hung.  I haven't been able to test that, because now I haven't
seen a hang or slowdown since I brought the second interface up.  

Could the idle interface have generated an interrupt which is never
cleared because the interface is down?  Or something similar?
-- 
Rob

Download attachment "signature.asc" of type "application/pgp-signature" (190 bytes)

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ