[j-nsp] BGP "Holdtime", " Active Holdtime" and "Preference" values

[Martin T]

Hi,
in case one has following settings active with it's BGP peer:

Holdtime: 90 Preference: 170
Active Holdtime: 90
Keepalive Interval: 30

..then what do they mean? As I understand, "Holdtime" is the maximum
number of seconds allowed to elapse between the time that a BGP system
receives successive keepalive or update messages from a peer. So if
the holdtime is configured to 90s, the "Holdtime" value under "show
bgp neighbor I.I.P.P" doesn't change, does it? How is "Active
Holdtime" different from "Holdtime"? And what does this "Preference"
mean?

Last but not least- in case of the configuration above, is it possible
that the connection between peers is down for example 80 seconds and
then comes back up, but as holdtime is set to 90s, the session stays
up?


regards,
martin

[Stefan Fouant]

Post by Martin T
Hi,
Holdtime: 90 Preference: 170
Active Holdtime: 90
Keepalive Interval: 30
..then what do they mean? As I understand, "Holdtime" is the maximum
number of seconds allowed to elapse between the time that a BGP system
receives successive keepalive or update messages from a peer. So if
the holdtime is configured to 90s, the "Holdtime" value under "show
bgp neighbor I.I.P.P" doesn't change, does it? How is "Active
Holdtime" different from "Holdtime"? And what does this "Preference"
mean?

Holdtime is the configured holdtime on the local device, whereas Active
Holdtime is the negotiated holdtime between the two peers, which should
be the minimum of the two peers holdtime configuration.

Preference is the value we assign to BGP routes learned from this
neighbor, which in this case is 170, the default. Preference is the
equivalent of Administrative Distance in Cisco, and allows the router to
determine which route to prefer should the identical route be learned
from multiple sources (ex: RIP[100] vs. BGP[170]).

Post by Martin T
Last but not least- in case of the configuration above, is it possible
that the connection between peers is down for example 80 seconds and
then comes back up, but as holdtime is set to 90s, the session stays
up?

Yes.

Stefan Fouant
JNCIE-ER, JNCIE-M, JNCIE-SEC, JNCI
Technical Trainer, Juniper Networks
http://www.shortestpathfirst.net
http://www.twitter.com/sfouant

[Martin T]

David, Stefan:
I configured "hold-time 20" and BGP session came up with an "Active
Holdtime: 20" as expected. Thank you for explanations!


regards,
martin

Post by Stefan Fouant

Hi,
? Holdtime: 90 Preference: 170
? Active Holdtime: 90
? Keepalive Interval: 30
..then what do they mean? As I understand, "Holdtime" is the maximum
number of seconds allowed to elapse between the time that a BGP system
receives successive keepalive or update messages from a peer. So if
the holdtime is configured to 90s, the "Holdtime" value under "show
bgp neighbor I.I.P.P" doesn't change, does it? How is "Active
Holdtime" different from "Holdtime"? And what does this "Preference"
mean?

Holdtime is the configured holdtime on the local device, whereas Active
Holdtime is the negotiated holdtime between the two peers, which should be
the minimum of the two peers holdtime configuration.
Preference is the value we assign to BGP routes learned from this neighbor,
which in this case is 170, the default. ?Preference is the equivalent of
Administrative Distance in Cisco, and allows the router to determine which
route to prefer should the identical route be learned from multiple sources
(ex: RIP[100] vs. BGP[170]).

Last but not least- in case of the configuration above, is it possible
that the connection between peers is down for example 80 seconds and
then comes back up, but as holdtime is set to 90s, the session stays
up?

Yes.
Stefan Fouant
JNCIE-ER, JNCIE-M, JNCIE-SEC, JNCI
Technical Trainer, Juniper Networks
http://www.shortestpathfirst.net
http://www.twitter.com/sfouant