Re: A few nasty bugs in the networking code

2001-03-21 Thread Garrett Wollman
< said: > I see. Thus a separate type is needed for the vlan interface, and > /usr/sbin/arp should be modified to comprehend the new arp-capable > interface type, shouldn't it? Aye. -GAWollman To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-net" in the body of the mes

Re: A few nasty bugs in the networking code

2001-03-21 Thread Yar Tikhiy
On Tue, Mar 20, 2001 at 03:09:59PM -0500, Garrett Wollman wrote: > < said: > > > Isn't it better to assign the IFT_ETHER type to the vlan interface? > > There might be other places in the code where vlans would behave > > unexpectedly because of their type... > > No, because SNMP and potentially

Re: A few nasty bugs in the networking code

2001-03-20 Thread Mike Tancsa
On 20 Mar 2001 15:10:25 -0500, in sentex.lists.freebsd.net you wrote: >< said: > >> Isn't it better to assign the IFT_ETHER type to the vlan interface? >> There might be other places in the code where vlans would behave >> unexpectedly because of their type... > >No, because SNMP and potentially

Re: A few nasty bugs in the networking code

2001-03-20 Thread Garrett Wollman
< said: > Isn't it better to assign the IFT_ETHER type to the vlan interface? > There might be other places in the code where vlans would behave > unexpectedly because of their type... No, because SNMP and potentially other network management utilities need to know about it. -GAWollman To Uns

Re: A few nasty bugs in the networking code

2001-03-20 Thread Yar Tikhiy
On Tue, Mar 20, 2001 at 07:05:58AM -0500, Mike Tancsa wrote: > > Do any of the VLAN patches fix the arp -d bug with VLAN interfaces ? > > i.e. arp -d does not work > > cbackup2# ifconfig vlan0 > vlan0: flags=8843 mtu 1500 > inet 192.168.112.1 netmask 0xff00 broadcast 192.168.112.255

Re: A few nasty bugs in the networking code

2001-03-20 Thread Mike Tancsa
On 19 Mar 2001 13:00:32 -0500, in sentex.lists.freebsd.net you wrote: >On Mon, Mar 19, 2001 at 06:32:44PM +0100, Luigi Rizzo wrote: >> >> > We are heading to a new release, but the bugs are still there. >> > >> > Could a commiter do me a favor and take a look at the following reports: >> >> wh

Re: A few nasty bugs in the networking code

2001-03-20 Thread Jeroen Ruigrok/Asmodai
-On [20010320 04:00], Josef Karthauser ([EMAIL PROTECTED]) wrote: [local gnats] >I do it just in case I need access to a PR. It's come in handy having >it lying around. Same here, and that coupled with X and tkgnats, lub it. -- Jeroen Ruigrok van der Werven/Asmodai .oUo. asmodai@[wxs.nl|free

Re: A few nasty bugs in the networking code

2001-03-19 Thread Josef Karthauser
On Tue, Mar 20, 2001 at 01:04:48AM +0100, Luigi Rizzo wrote: > > Actually, I think quoting PR#s is a more than acceptable way of > > pointing things out. They're very easy to look up for anyone (and > > committers get the extra advantage of using query-pr on freefall) and > > IF you have connect

Re: A few nasty bugs in the networking code

2001-03-19 Thread Bruce Evans
On Tue, 20 Mar 2001, Luigi Rizzo wrote: > > Actually, I think quoting PR#s is a more than acceptable way of > > pointing things out. They're very easy to look up for anyone (and > > committers get the extra advantage of using query-pr on freefall) and > > IF you have connectivity while you are

Re: A few nasty bugs in the networking code

2001-03-19 Thread Luigi Rizzo
> Actually, I think quoting PR#s is a more than acceptable way of > pointing things out. They're very easy to look up for anyone (and > committers get the extra advantage of using query-pr on freefall) and IF you have connectivity while you are reading, which is my whole point. The one big advan

Re: A few nasty bugs in the networking code

2001-03-19 Thread Yar Tikhiy
Hello Garrett, On Mon, Mar 19, 2001 at 01:08:32PM -0500, Garrett Wollman wrote: > > I have taken a look at all of these and your suggested fixes appear to > be correct in concept. I have not tested any of them, however. As for me, I can see a fixed system work perfectly for months. It was an un

Re: A few nasty bugs in the networking code

2001-03-19 Thread Jordan Hubbard
Actually, I think quoting PR#s is a more than acceptable way of pointing things out. They're very easy to look up for anyone (and committers get the extra advantage of using query-pr on freefall) and it sure beats wearing one's fingers out by entering the same information over and over again. We

Re: A few nasty bugs in the networking code

2001-03-19 Thread Garrett Wollman
< said: > if_delmulti() (net/if.c) does not notify a corresponding interface > driver when a _link-layer_ multicast group is being left. > There is a mtod() without a prior m_pullup() in netinet/if_ether.c. > The system might be likely to crash sometimes... > The vlan driver don't update byte/p

Re: A few nasty bugs in the networking code

2001-03-19 Thread Yar Tikhiy
On Mon, Mar 19, 2001 at 06:32:44PM +0100, Luigi Rizzo wrote: > > > We are heading to a new release, but the bugs are still there. > > > > Could a commiter do me a favor and take a look at the following reports: > > which are about ??? you know we are better at parsing text strings than > number

Re: A few nasty bugs in the networking code

2001-03-19 Thread Luigi Rizzo
> Hi there, > > Once upon a time I ran into several bugs in the FreeBSD networking > code. Being a humble FreeBSD user, I started send-pr and wrote bug > reports including detailed descriptions and fixes on all of them, > but they still seem to remain unnoticed by the responsible. > We are headi