1

I have several Cisco 2960 switches here, all configured for RSTP.

15.0(2)SE6
C2960-LANBASEK9-M

spanning-tree mode rapid-pvst

Topology is as follows:

S1 -- S2 -- S4
  \        /
   \__S3__/

I change the bridge ID on S3 so that it will become root bridge.

I then have the following debug output:

Apr  7 12:25:27.173: RSTP(1): updt roles, received superior bpdu on Gi0/2
Apr  7 12:25:27.173: RSTP(1): Gi0/1 is now root port
Apr  7 12:25:27.173: RSTP(1): Gi0/2 blocked by re-root
Apr  7 12:25:27.173: RSTP(1): Gi0/2 is now designated
Apr  7 12:25:27.182: STP[1]: Generating TC trap for port GigabitEthernet0/1
Apr  7 12:25:27.182: RSTP(1): updt roles, received superior bpdu on Gi0/1
Apr  7 12:25:27.182: RSTP(1): we become the root bridge
Apr  7 12:25:27.182: RSTP(1): Gi0/1 is now designated
Apr  7 12:25:27.182: RSTP(1): transmitting a proposal on Gi0/2
Apr  7 12:25:27.190: RSTP(1): received an agreement on Gi0/2
Apr  7 12:25:27.199: STP[1]: Generating TC trap for port GigabitEthernet0/2

At which point in this log is the RSTP process finished and will start forwarding frames again? My thought is at 27.190, a total of 13 ms. Is this assumption correct or is the last line in the log still part of the RSTP process?

  • 1
    On which port? I think you mean to ask when the switches have all converged. STP, including RSTP` is an ongoing process. BPDUs are being sent and received all the time. You will begin forwarding on G0/2 when it becomes a designated port at 12:25:27.173, and you will begin forwarding on G0/1 when it becomes a designated port at 12:25:27.182 None of that means the switches on the other end of those links aren't still blocking for a few more milliseconds, which appears to be the case for the switch connected to G0/2. – Ron Maupin Apr 22 '16 at 12:53
  • On both g0/2 and g0/1. g0/2 becomes designated at .173, I am confused about the last line at .199. No more debug output is produced afterwards. I have not tried syncing debug output for the other switches. – user3549596 Apr 22 '16 at 13:58
  • The last line looks like it is generating an SNMP trap for the topology change, as does the 5th line. – Ron Maupin Apr 22 '16 at 14:27
  • That seems to make sense. So actual convergence on S3 would be completed at .190? At which point in time the other switches are finished is not relevant right now. – user3549596 Apr 23 '16 at 14:23
  • Convergence is when all the switches agree. You don't have enough information to determine when that happens. S3 thinks it's done at 12:25:27.182 when both its interfaces are correctly set, and it will begin sending traffic on both of them. That doesn't mean that the other switches won't block the traffic until they have finished running the STP algorithm. When all switches have finished, you are converged. Each switch can finish at a different time. Some may be done before S3 (S2 may be done first), and some may be done after S3 (whichever switch is connected to G0/2 of S3). – Ron Maupin Apr 23 '16 at 14:30
0

I think you mean to ask when the switches have all converged. STP, including RSTP` is an ongoing process. BPDUs are being sent and received all the time.

You will begin forwarding on G0/2 when it becomes a designated port at 12:25:27.173, and you will begin forwarding on G0/1 when it becomes a designated port at 12:25:27.182 None of that means the switches on the other end of those links aren't still blocking for a few more milliseconds, which appears to be the case for the switch connected to G0/2.

Without looking at the other switches, you cannot tell when they have all converged, which may take just a little longer than S3. With what you have provided, you can't tell when they are all in agreement.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.