netfilter buglog - Jun 2003 - [Bug 47] conntrack breaks nfs, corrupted packets

https://bugzilla.netfilter.org/cgi-bin/bugzilla/show_bug.cgi?id=47





------- Additional Comments From dg@ezcom.de  2003-06-30 16:17 -------
today i replaced my 8139too with a 3c59x and the corruptions are gone



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.netfilter.org/cgi-bin/bugzilla/show_bug.cgi?id=47

laforge@netfilter.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |WORKSFORME



------- Additional Comments From laforge@netfilter.org  2003-06-30 16:52 -------
since this seems to be a confirmed bug on one particular NIC chipset, I think
it's unlikely to be a bug in the generic netfilter core.

Please feel free to reopen once there is new evidence :)



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.netfilter.org/cgi-bin/bugzilla/show_bug.cgi?id=47





------- Additional Comments From kaber@trash.net  2003-07-06 01:52 -------
Bug status can be changed to invalid, i managed to reproduce the ill 
behaviour without anything of netfilter compiled in.

The malformed packets can be triggered by cating something over a ssh 
connection while transfering data at high speed, it "feels" like the
card is
getting packets too fast and corrupts them. i believe the nfs + ip_conntrack
combination has similar effects like cat because ip_conntrack causes bursts
in the nfs case because the inter-packet spacing is lost due to defragmentation.
non-bursty traffic doesn't cause this effect.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.netfilter.org/cgi-bin/bugzilla/show_bug.cgi?id=47

laforge@netfilter.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|WORKSFORME                  |



------- Additional Comments From laforge@netfilter.org  2003-07-06 13:34 -------
bug can now be closed as invalid



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.netfilter.org/cgi-bin/bugzilla/show_bug.cgi?id=47

laforge@netfilter.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|REOPENED                    |RESOLVED
         Resolution|                            |INVALID



------- Additional Comments From laforge@netfilter.org  2003-07-06 13:35 -------
see kaber's last comment.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.netfilter.org/cgi-bin/bugzilla/show_bug.cgi?id=47





------- Additional Comments From gandalf@netfilter.org  2003-07-06 17:58 -------
Patrick, It would be great if you could report this problem to
http://bugzilla.kernel.org/ so it has a chance of getting fixed.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.netfilter.org/cgi-bin/bugzilla/show_bug.cgi?id=47





------- Additional Comments From kaber@trash.net  2003-07-06 20:10 -------
this is probably not necessary, with PIO i can't reproduce the problem at
all
(the help text mentions problems with motherboards with memory-consistency
issues).
i'm going to test the card on a different, much newer board. if it turns out
ok
there is no need for a bugreport.



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.