I have been attempting to test the -mm kernels. I was having problems
with the anticipatory scheduler up causing the machine to hang during
loading of the USB driver. The recent fixes that went into -mm8 seem to
have corrected this problem, unfortunately I'm now hitting another
issue.
It seems that -mm8 breaks MASQ support in iptables. Actually I can't
get any MASQ/NAT to work at all. The same exact config works fine with
vanilla 2.5.64 and 2.5.64-ac4.
Backing out the brlock patches returns everything to normal operation so
there's something happening with those. I'll try to look at them
individually if I get a chance but I know practically nothing about that
code so somebody will probably spot a problem quicker.
Later,
Tom
This actually broke in -mm7, but I don't know what causes it.
I have to admit, I haven't even looked at the patch to see what changed.
Oh well, I suspect good ol' 65-mm1 will fix things up. If so, my TiVo
could stop holding it's breath. ;)
Anyone else seeing this?
On Mon, 2003-03-17 at 11:29, Tom Sightler wrote:
> I have been attempting to test the -mm kernels. I was having problems
> with the anticipatory scheduler up causing the machine to hang during
> loading of the USB driver. The recent fixes that went into -mm8 seem to
> have corrected this problem, unfortunately I'm now hitting another
> issue.
>
> It seems that -mm8 breaks MASQ support in iptables. Actually I can't
> get any MASQ/NAT to work at all. The same exact config works fine with
> vanilla 2.5.64 and 2.5.64-ac4.
>
> Backing out the brlock patches returns everything to normal operation so
> there's something happening with those. I'll try to look at them
> individually if I get a chance but I know practically nothing about that
> code so somebody will probably spot a problem quicker.
>
> Later,
> Tom
>
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
Shawn <[email protected]> wrote:
>
> This actually broke in -mm7, but I don't know what causes it.
>
> I have to admit, I haven't even looked at the patch to see what changed.
> Oh well, I suspect good ol' 65-mm1 will fix things up. If so, my TiVo
> could stop holding it's breath. ;)
>
> Anyone else seeing this?
Stephen is looking into it. Please send him your iptables
configuration info.
On Tue, 2003-03-18 at 05:55, Andrew Morton wrote:
> Shawn <[email protected]> wrote:
> >
> > This actually broke in -mm7, but I don't know what causes it.
> >
> > I have to admit, I haven't even looked at the patch to see what changed.
> > Oh well, I suspect good ol' 65-mm1 will fix things up. If so, my TiVo
> > could stop holding it's breath. ;)
> >
> > Anyone else seeing this?
>
> Stephen is looking into it. Please send him your iptables
> configuration info.
My iptables config is just a simple one-liner as follows:
iptables -t nat -A POSTROUTING -o eth1 -j MASQUERADE
eth1 is an Aironet PCCARD wireless adapter connected to my corporate
network
eth0 is 3c556 (3c59x driver) Mini-PCI 10/100 with various systems
connected to it.
I also regularly NAT my VMware virtual machine (normally configured for
host only networking) which sits on vmnet1.
I've tried several variations of the iptables option, such as:
iptables -t nat -A POSTROUTING -s 192.168.4.1/24 -j MASQUERADE
iptables -t nat -A POSTROUTING -o ! eth0 -j MASQUERADE
All of these variations work without the brlock patches backed out.
Actually, I haven't been able to get any MASQ/NAT options to work with
the brlock removal patches.
Let me know if you need more info.
Later,
Tom
Shawn wrote:
>This actually broke in -mm7, but I don't know what causes it.
>
>I have to admit, I haven't even looked at the patch to see what changed.
>Oh well, I suspect good ol' 65-mm1 will fix things up. If so, my TiVo
>could stop holding it's breath. ;)
>
>Anyone else seeing this?
>
>
I'm seeing a weird problem with later 2.5.64-mm,
probably related - when the problem hits, I have
to "make things work" quickly, so there is not a
lot of time to analyze (booting back into the
latest -ac kernel makes everyone happy)
The box in question, running RH 8.0, is set up
as a dhcp server, iptables firewall, and squid
proxy for a small network of 6 systems (3 linux,
3 ms windows)
After booting up, it all works initially, but after
some time, the windows/dhcp users complain
that they can't get out to the internet anymore.
Very odd, since my internet access via a linux
workstation seems fine - I could just say "wow,
it sure sucks to be a win doze user" and move
on, but I can't quite get away with that...
So the problem is rather interesting, and multi-
faceted - I'm not sure at this point whether the
problems follow the ms windows users, or the
dhcp users, or those not plugged into the same
switch that's in my office... but I'm willing to
try possible fixes through a brute force trial
and error approach.
Joe