Mailing List Archive

On Mon, 22 Dec 2003, Greg Troxel wrote:

> It might be time for 0.96.5; a number of things are fixed.

And what about my patches? ;-)

Best regards,

Krzysztof Olêdzki

[pointer to quagga-dev 531]

I committed the patch in 438 (drop privs on error case in bgpd); it
compiles (but I didn't test it).

In 439, I am not sure of the consequences of the change, in terms of
whether the other daemons are set up to handle this and what existing
practice is for ZAPI_MESSAGE_DISTANCE. I didn't understand from your
message which of them you checked and/or tested.

I don't run rip at all, so I would like to avoid committing to it.

On Tue, 23 Dec 2003, Greg Troxel wrote:

> [pointer to quagga-dev 531]
>
> I committed the patch in 438 (drop privs on error case in bgpd); it
> compiles (but I didn't test it).

OK :)

> In 439, I am not sure of the consequences of the change, in terms of
> whether the other daemons are set up to handle this and what existing
> practice is for ZAPI_MESSAGE_DISTANCE. I didn't understand from your
> message which of them you checked and/or tested.

IMHO they are:

ripd/rip_zebra.c:
(...)
if (CHECK_FLAG (api.message, ZAPI_MESSAGE_DISTANCE))
api.distance = stream_getc (s);
(...)


ripngd/ripng_zebra.c:
(...)
if (CHECK_FLAG (api.message, ZAPI_MESSAGE_DISTANCE))
api.distance = stream_getc (s);
else
api.distance = 0;
(...)


ospfd/ospf_zebra.c:
(...)
if (CHECK_FLAG (api.message, ZAPI_MESSAGE_DISTANCE))
api.distance = stream_getc (s);
(...)


ospf6d/ospf6_zebra.c:
(...)
if (CHECK_FLAG (api.message, ZAPI_MESSAGE_DISTANCE))
api.distance = stream_getc (s);
else
api.distance = 0;
(...)


bgpd/bgp_zebra.c:
(...)
if (CHECK_FLAG (api.message, ZAPI_MESSAGE_DISTANCE))
api.distance = stream_getc (s);
(...)


isisd/isis_zebra.c:
if (CHECK_FLAG (api.message, ZAPI_MESSAGE_DISTANCE))
api.distance = stream_getc (stream);


> I don't run rip at all, so I would like to avoid committing to it.
OK.


Best regards,

Krzysztof Olêdzki

On Tue, 23 Dec 2003, Greg Troxel wrote:

<CIACH>
> In 439, I am not sure of the consequences of the change, in terms of
> whether the other daemons are set up to handle this and what existing
> practice is for ZAPI_MESSAGE_DISTANCE. I didn't understand from your
> message which of them you checked and/or tested.
Ah, one more thing - I have been running zebrad+ripd+bgpd for about two
weeks with this change.

Best regards,

Krzysztof Olêdzki

On Tue, 23 Dec 2003, Greg Troxel wrote:

> In 439, I am not sure of the consequences of the change, in terms
> of whether the other daemons are set up to handle this and what
> existing practice is for ZAPI_MESSAGE_DISTANCE.

It seems the protocol daemons send this, but zebra itself does not
send distance for routes when it redistributes. Although, at least
ospfd and ripd do check whether DISTANCE is included in the message.
(ospfd doesnt seem to use it though, AFAICT).

> I don't run rip at all, so I would like to avoid committing to it.

Its something to test out though. The daemons I've looked at clearly
do expect zebra to possibly send distance information.

regards,
--
Paul Jakma paul@clubi.ie paul@jakma.org Key ID: 64A2FF6A
warning: do not ever send email to spam@dishone.st
Fortune:
What the scientists have in their briefcases is terrifying.
-- Nikita Khruschev