headers
jean-mickael guerin | 1 Apr 2004 20:15

Possible crash in ip_reass_ttl_decr()

Hello,

I got a kernel panic in ip_reass_ttl_decr() when my NetBSD-1.6ZI 
receives IP fragments. I think it is related to the loop
for (i = IPFRAGTTL, median = 0; i >= 0; i--) {

where i was declared unsigned. I use following fix that seems to work:

<       u_int i, nfrags, median;
---
 >       int i;
 >       u_int nfrags, median;

Jean-Mickael
Permalink | Reply | 0 comments |
headers
Hanspeter Roth | 2 Apr 2004 21:14
Picon
Favicon

userppp and loopback on tun0

Hello,

I had tried the userppp from pkgsrc but didn't succeed. So I
installed one from http://www.awfulhak.org/ppp.html (ppp-030706).
With this one I'm able to establish a single connection via isdnd.

ifconfig of tun0 is:

tun0: flags=80f5<UP,DEBUG,POINTOPOINT,NOTRAILERS,RUNNING,NOARP,MULTICAST> mtu 1500
	inet 10.0.0.1 -> 10.0.0.2 netmask 0xfffffffc

If I'm trying to establish a setup with bundled channels tun0 gets
the loopback flag:

tun0: flags=a079<UP,LOOPBACK,POINTOPOINT,NOTRAILERS,RUNNING,LINK1,MULTICAST> mtu 1500
	inet 127.0.0.1 -> 10.0.0.2 netmask 0xffffff00

The entry in ppp.conf is:

 set cd 20
 set phone 100000
 set device /dev/isdnbchan0 /dev/isdnbchan1
 set login
 set authname name
 set authkey key
 set timeout 178
 set ifaddr 10.0.0.1/0 10.0.0.2/0 255.255.255.0 0.0.0.0
 disable loopback

 set mrru 1524				# Multilink mode please
(Continue reading)

Permalink | Reply | 0 comments |
headers
Pavel Cahyna | 2 Apr 2004 22:10
Picon

Re: Slow NFS transfer start

Hello,

On Wed, 31 Mar 2004 19:10:58 +0200, Arto Huusko wrote:

> I have the following problem: when an NFS client connects to NFS server
> and starts moving large chunks of data from the server to the client,
> the NFS session seems to hang for a long while doing nothing, until it

Have you tried to ping from the client to the server at this moment? I recall
having a similar (but probably not the same) problem and the transfer which was
hanging restarted. (But after a while it stopped again...)

Bye	Pavel
Permalink | Reply | 0 comments |
headers
Jason Thorpe | 3 Apr 2004 01:03

Re: Ethernet flowcontrol patch for 1.6.2


On Mar 15, 2004, at 7:19 AM, Jason Thorpe wrote:

>
> On Mar 14, 2004, at 12:18 PM, HITOSHI Osada wrote:
>
>> The patch is located
>> http://homepage3.nifty.com/hosada/NetBSD/ether-flowcontrol/index.html
>>
>> Comments, Suggestions are welcome.
>
> Thank you.  I have downloaded your patch, and will take a look at it 
> over the next few days.

Sorry for the delay.  I've looked over your patch, and I think you 
forgot to include the patch for if_media.h?  If so, would you be so 
kind as to send that to me?

Thanks!

         -- Jason R. Thorpe <thorpej <at> wasabisystems.com>
Permalink | Reply | 2 comments |
headers
HITOSHI Osada | 3 Apr 2004 09:35
Favicon

Re: Ethernet flowcontrol patch for 1.6.2

At Fri, 2 Apr 2004 15:03:12 -0800,
Jason Thorpe <thorpej <at> wasabisystems.com> wrote:
> 
> [1  <text/plain; US-ASCII (7bit)>]
> 
> On Mar 15, 2004, at 7:19 AM, Jason Thorpe wrote:
> 
> >
> > On Mar 14, 2004, at 12:18 PM, HITOSHI Osada wrote:
> >
> >> The patch is located
> >> http://homepage3.nifty.com/hosada/NetBSD/ether-flowcontrol/index.html
> >>
> >> Comments, Suggestions are welcome.
> >
> > Thank you.  I have downloaded your patch, and will take a look at it 
> > over the next few days.
> 
> Sorry for the delay.  I've looked over your patch, and I think you 
> forgot to include the patch for if_media.h?  If so, would you be so 
> kind as to send that to me?

  I've updated my patch(20040317). Please download from my homepage.
Permalink | Reply | 0 comments |
headers
Sorin Cioara | 5 Apr 2004 08:32

Creating a GRE interface wedges my machine

Hi,

(First post here, so please be gentle...)

First, I have whole bunch of identical boxes that run the same level 
OS and are rock solid - supermicro 1U, P4 servers running NetBSD 1.6.2.

I am trying to create a GRE tunnel between two of the boxes, slightly 
different hardware. Although I have been unsuccessful at actually 
passing data through the tunnel, I seem to have been able to create 
the tunnel. At least, according to "ifconfig" the interface is up, and 
"netstat" shows routes.

The problem is, that while I was trying to figure out why packets were 
escaping into the ether never to return, I managed to wedge the 
machines a few times. I can't reproduce the sequence every time, but 
basically when I issued the
"ifconfig gre0 create inet 10.0.0.1 10.0.0.3 netmask 255.255.255.192 
up" command the machine just locked. It gave NO sign of life; it 
didn't even respond to keyboard input. In fact, the signal to the 
monitor was even cut off - monitor went into power saving mode. At 
first I thought it could be a hardware issue, but it happened on both 
machines I was trying it on. In fact, once the command completed OK on 
machine "A" and when I switched to an tried it on machine "B" machine 
"A" wedged.

I am at the end of my, obviously limited. wits here; and I am hoping 
someone can shed some light on this for me.

By the way, I do have a GRE tunnel working perfectly between two other
(Continue reading)

Permalink | Reply | 1 comment |
headers
Mihai CHELARU | 5 Apr 2004 11:28

Secondary addresses

Hi,

I have an alias ip addresses issue on 1.6.2. Here is what it's 
configured:

10.10.0.2/24 as primary address on tlp0
10.12.0.2/24 as secondary(alias) address on tlp0

I've made some tests in order to see the source address used to reach 
10.12.0.1 (on the alias segment). Although the route is directly 
connected to this interface via its alias, the alias is never used to 
reach 10.12.0.1 but, instead, 10.10.0.2 is used (even for ping -I 
10.12.0.2 and traceroute -s 10.12.0.2). Is this the normal behaivior ? 
IMHO, the alias address should be used as 10.10.0.2 should connect to 
10.12.0.0/24 only via default-gateway.

--

-- 
Any oppinion is appreciated,
M
Permalink | Reply | 0 comments |
headers
Martin Husemann | 5 Apr 2004 12:10
Picon

Re: Creating a GRE interface wedges my machine

On Mon, Apr 05, 2004 at 03:32:08PM +0900, Sorin Cioara wrote:
> (First post here, so please be gentle...)

We always are - aren't we?

> The problem is, that while I was trying to figure out why packets were 
> escaping into the ether never to return, I managed to wedge the 
> machines a few times. I can't reproduce the sequence every time, but 
> basically when I issued the
> "ifconfig gre0 create inet 10.0.0.1 10.0.0.3 netmask 255.255.255.192 
> up" command the machine just locked.

Could this be a similar effect than the endless recursion described
in this PR?

http://www.netbsd.org/cgi-bin/query-pr-single.pl?number=21483

Martin
Permalink | Reply | 0 comments |
headers
Nitin P Mahajan | 6 Apr 2004 17:33

Ethernet receive testing.

Hi!

	I want to test the receive part of my Ethernet driver. My
platform doesn't have a GUI and display adapter.

Is there any other tool other than ethereal with a command line
interface with which I can read the incoming and out going packets like
ethereal does?

Regards

-Nitin
Permalink | Reply | 3 comments |
headers
cube | 6 Apr 2004 17:36

Re: Ethernet receive testing.

On Tue, Apr 06, 2004 at 09:03:33PM +0530, Nitin P Mahajan wrote:
> 	I want to test the receive part of my Ethernet driver. My
> platform doesn't have a GUI and display adapter.
> 
> Is there any other tool other than ethereal with a command line
> interface with which I can read the incoming and out going packets like
> ethereal does?

tcpdump(8)

Also, you should note that ethereal installs 'tethereal', which serves
the same purpose.

Quentin Garnier.
Permalink | Reply | 1 comment |

Gmane