On Nov 9, 2007 7:46 PM, Mike Silbersack <[EMAIL PROTECTED]> wrote:
>
> On Fri, 9 Nov 2007, Matt Reimer wrote:
>
> > On a eight core machine running RELENG_7 I'm seeing TCP stalls,
> > sometimes lasting up to 60 seconds or so. While trying to track this
> > down I noticed that net.inet.tcp.syncache.
On Fri, 9 Nov 2007, Matt Reimer wrote:
On a eight core machine running RELENG_7 I'm seeing TCP stalls,
sometimes lasting up to 60 seconds or so. While trying to track this
down I noticed that net.inet.tcp.syncache.count is negative. Should it
be possible for the count to go negative? Perhaps it
On a eight core machine running RELENG_7 I'm seeing TCP stalls,
sometimes lasting up to 60 seconds or so. While trying to track this
down I noticed that net.inet.tcp.syncache.count is negative. Should it
be possible for the count to go negative? Perhaps it indicates a race,
or the counter is wrongl
I suppose output of "uname -a", "pciconf -lv" and "dmesg" would be useful.
Good time of the day.
I'm not pleased to emphasize that BGE makes FreeBSD 7.0/8.0 systems
completely unusable on machines with BCM5787 (NetLink BCM5787M Gigabit
Ethernet)
Install CD hangs as soon as it tries to load
Good time of the day.
I'm not pleased to emphasize that BGE makes FreeBSD 7.0/8.0 systems
completely unusable on machines with BCM5787 (NetLink BCM5787M Gigabit
Ethernet)
Install CD hangs as soon as it tries to load driver for Ethernet device
(tested on 7-CURRENT (snapshots 200708, 200709, 2
On Friday 09 November 2007, Dag-Erling Smørgrav wrote:
> Max Laier <[EMAIL PROTECTED]> writes:
> > No, I don't see why these two should behave differently, but you
> > should add a "scrub in on sk0" in any case.
>
> scrub is known and documented to interfere with NFS.
Only with broken NFS clients
Max Laier <[EMAIL PROTECTED]> writes:
> No, I don't see why these two should behave differently, but you should
> add a "scrub in on sk0" in any case.
scrub is known and documented to interfere with NFS.
DES
--
Dag-Erling Smørgrav - [EMAIL PROTECTED]
_
