show spanning-tree detail counstatistics

[Steph1963]

Hi,

Is there anybody who know the meaning of forward delay, and hold in the show spanning-tree detail

Port 24 (FastEthernet0/24) of VLAN0001 is forwarding

Port path cost 19, Port priority 128, Port Identifier 128.24.

Designated root has priority 1, address 000b.46eb.2fc0

Designated bridge has priority 32768, address 001e.7adf.faa8

Designated port id is 128.4, designated path cost 1321

Timers: message age 3, forward delay 0, hold 0

Number of transitions to forwarding state: 1

Link type is point-to-point by default

BPDU: sent 3, received 108

Max age: Does this field is locally calculated on each switch by indicating the time elapse since the last BPDU receive, or does it used the age field in the BPDU and used max-age - age to calculate the max age.

Does forward delay corresponds to the time the switch is listening for better BPDU or does is corresponds to the learning period where the switch update it`s MAC table or is it corresponding to both? In other words, does the forward delay varies between 15 and 0 or between 30 an0 sec?

Hold: I was never able to find any information on that staistics.

Thanks for your help

Stephane

[lgijssel · Accepted solution]

Max age: is the one as advertized by the root. This overrides any local value.

Forward delay: This corresponds to the time the switch is listening for a better BPDU. Typically the learning phase also lasts 1 x FWD_DLY.

Hold time: The time interval (in seconds) during which no more than two Configuration BPDU's (Bridge Protocol Data Unit) shall be transmitted by the switch. (Google is your friend. )

regards,

Leo

[Steph1963]

Hi Leo,

Thanks for your help,

Can you just confirm that the message age  carried within the BPDU corresponds to the max age substract by one each time BPDU is going through a switch and the message age time display by a show spanning-tree detail is an active timer that vary between the message age in the BPDU and 0. Can we say that a message age to 0 would indicate to a switch that it`s root is no more longer valid.

Would it be possible that hold time on 2950 and 2960 is not configurable, that might explain why there is not much information on this for these two type of switches.

Thanks again

Stéphane

[lgijssel]

Looking at the counters, I think we are running rapid stp here. This operates slightly different from conventional stp.

Please check the link below:

http://www.cisco.com/en/US/tech/tk389/tk621/technologies_white_paper09186a0080094cfa.shtml#topic4

All bridges transmit bpdu's instead of relaying the root bpdu. This is where the 'odd' value of 3s comes from.

The port details you posted make me think also that it was in portfast mode. (FWD_DLY = 0)

regards,

Leo

[Steph1963]

Hi Leo,

These results are coming from spanning tree (802.1D) so that probably explain why the message age is 3. I would assume that age could go up to max age (for a 2 switch back to back) and only goes up to 2 for a rapid spanning-tree (802.1W)

Do not know if we can assume that forward delay  is set to 0 when the switch is not in listening state.

Thanks

Stéphane