Mailing List Archive

On Thu, May 13, 2004 at 01:51:30PM -0400, Andrew J. Schorr wrote:
> I am attaching an updated version of the patch that applies cleanly
> against the latest CVS snapshot.

I found a bug in my patch over the weekend (in two zlog_warn messages
in ripd/rip_interface.c), so I am attaching a small fix (that should be
applied after the main patch). Or I can resubmit the patch with the fix
incorporated.

Is there any further testing I should be doing? We are currently
running ospfd with that patch in production, and it seems to work for us.
I also realize that my patch does not include ChangeLog updates; I can
add that if it would be helpful, I just didn't want to spend the time
until it was clear what was going to be committed.

Thanks,
Andy

On Mon, 17 May 2004, Andrew J. Schorr wrote:

> I found a bug in my patch over the weekend (in two zlog_warn
> messages in ripd/rip_interface.c), so I am attaching a small fix
> (that should be applied after the main patch). Or I can resubmit
> the patch with the fix incorporated.
>
> Is there any further testing I should be doing?

ripd, bgpd? :)

Also, I'd like to see the router-lsa's produced for the both the
cases of a local/peer addressed and a address/broadcast addressed PtP
link.

> We are currently running ospfd with that patch in production, and
> it seems to work for us. I also realize that my patch does not
> include ChangeLog updates; I can add that if it would be helpful, I
> just didn't want to spend the time until it was clear what was
> going to be committed.

I think it's worth having, it will be committed at some stage.
there's more 'stable' patches to go in first though.

ChangeLog entries would be very useful too.

> Thanks,
> Andy

regards,
--
Paul Jakma paul@clubi.ie paul@jakma.org Key ID: 64A2FF6A
warning: do not ever send email to spam@dishone.st
Fortune:
God gives us relatives; thank goodness we can chose our friends.

On Mon, May 17, 2004 at 04:07:17PM +0100, Paul Jakma wrote:
> On Mon, 17 May 2004, Andrew J. Schorr wrote:
> > Is there any further testing I should be doing?
>
> ripd, bgpd? :)

I already tested those (as mentioned in my post [quagga-dev 1175] on
Thursday). I found no problems. Plus /31 subnets also work with the patch.

> Also, I'd like to see the router-lsa's produced for the both the
> cases of a local/peer addressed and a address/broadcast addressed PtP
> link.

How do I get that info for you? Is that in the output of "debug ospf lsa"?

> I think it's worth having, it will be committed at some stage.
> there's more 'stable' patches to go in first though.

OK, great. There are a couple of other ospfd problems/enhancements
that I'd like to patch, but I guess I'll wait until after this
is committed and the code stabilizes.

> ChangeLog entries would be very useful too.

OK, I'll do that when I get a chance.

-Andy

On Mon, 17 May 2004, Andrew J. Schorr wrote:

> How do I get that info for you? Is that in the output of "debug ospf lsa"?

show ip os da rou <LSA ID>

should do fine. (on originating router and a neighbour would be nice
- unless they are identical, dont bother then ;) )

'show interface <ptp iface>' would be useful too.

> OK, great. There are a couple of other ospfd problems/enhancements
> that I'd like to patch, but I guess I'll wait until after this is
> committed and the code stabilizes.

Sure.

> OK, I'll do that when I get a chance.
>
> -Andy

regards,
--
Paul Jakma paul@clubi.ie paul@jakma.org Key ID: 64A2FF6A
warning: do not ever send email to spam@dishone.st
Fortune:
Don't talk to me about naval tradition. It's nothing but rum, sodomy and
the lash.
-- Winston Churchill

On Mon, May 17, 2004 at 04:51:51PM +0100, Paul Jakma wrote:
> show ip os da rou <LSA ID>
>
> should do fine. (on originating router and a neighbour would be nice
> - unless they are identical, dont bother then ;) )
>
> 'show interface <ptp iface>' would be useful too.
>

OK, test results are attached. In both cases, I have two routers
connected by a PtP link. Each is also advertising an attached
ethernet network to the PtP peer (but there are no neighbors
on the ethernets, those interfaces are passive).

In the first case, I used a dedicated 192.168.162.0/31 subnet.
The local ethernet is 192.168.11.0/24, and the remote is 192.168.130/24.

In the second case, I borrowed IP addresses for the PtP link:
192.168.11.62/32 on the local side, and 192.168.130.3/32 on the
remote side (addresses borrowed from the ethernet interfaces).

I dumped out the database and interface info on both sides of the
link in each case, but I am only attaching the output from the
local side, since the remote side was essentially identical (other
than the obvious differences). I can send those also if you would
like to see them.

I hope this helps. Please let me know if you would like to see any other
test results.

Thanks,
Andy