ExtremeSwitching (Other)

 View Only
  • 1.  Topology change every 1h

    Posted 09-01-2016 07:16
    Hello together,
    we have a big problem on our network, topology change takes place every hour and in the morning sometimes every 3 minutes.
    We use RSTP and BPDU on all edge Ports.
    We have 2 Core S8 and about 50 Stacks with B2, B3 and B5, altogether about 150 Enterasys switches.
    Both S8 are with 2x10Gbit connected together to a lag.0.1 with all vlans



    First S8 is the Root Bridge

    Designated Root Priority - 4096
    Designated Root Cost - 0
    Designated Root Port - 0
    Root Max Age - 20 sec
    Root Hello Time - 2 sec
    Root Forward Delay - 15 sec
    Bridge ID MAC Address - xx
    Bridge ID Priority - 4096
    Bridge Max Age - 20 sec
    Bridge Hello Time - 2 sec
    Bridge Forward Delay - 15 sec
    Topology Change Count - 24778
    Time Since Top Change - 00 days 00:35:02
    Max Hops - 20

    Second S8

    Designated Root Priority - 4096
    Designated Root Cost - 1
    Designated Root Port - lag.0.1
    Root Max Age - 20 sec
    Root Hello Time - 2 sec
    Root Forward Delay - 15 sec
    Bridge ID MAC Address - xx
    Bridge ID Priority - 8192
    Bridge Max Age - 20 sec
    Bridge Hello Time - 2 sec
    Bridge Forward Delay - 15 sec
    Topology Change Count - 24755
    Time Since Top Change - 00 days 00:35:02
    Max Hops - 20

    Our Problem is that the Topology change doesn´t match all Stacks, there are only 17/50 and both Cores which are doing that. We disabled the complete redundancy and it still happens even there where only one stack is connected to both Cores...

    How can we find out who causes this Topology change?

    All other topics here didn´t help us.

    Thanks for help
    greets



  • 2.  RE: Topology change every 1h

    Posted 09-01-2016 07:37
    I am surpris why stp or rstp still in used if u use summitstack or bd switches... The reason why stp or rstp is disable by default. All switch to switch link should be IEEE 802.1Q. No daisy chain. To prevent looping, u should use elrp. To load share, u should use IEEE 802.3ad...


  • 3.  RE: Topology change every 1h

    Posted 09-01-2016 07:37
    He has stated that he is using Enterasys S8, B-Series Switches. They do not have ELRP. Only the Extreme Branded (Purple) switches have ELRP.


  • 4.  RE: Topology change every 1h

    Posted 09-01-2016 11:43
    Hi Adam,

    the GTAC Knowledge site has a few articles regarding STP topology change issues:
    Perhaps one of those can help you identifying the device causing topolgy changes.

    HTH,
    Erik


  • 5.  RE: Topology change every 1h

    Posted 09-02-2016 15:53
    Adam,

    The articles Erik pointed out should help. If you need more assistance give us a call at the GTAC.



  • 6.  RE: Topology change every 1h

    Posted 09-12-2016 11:40
    first at all thanks for your help and usefull hints.
    I found some interesting things...

    when Core 1 shows topology change
    <166>Sep 12 14:57:00 X.X.X.X Spantree[4]Topology Change: RECEIVED on SID = 0
    Port = lag.0.1
    on Core 2 there is nothing at the same time.

    And when Core 2 shows topology change
    <166>Sep 12 14:22:04 X.X.X.X Spantree[4]Topology Change: RECEIVED on SID = 0
    Port = lag.0.1
    An core 1 was nothing at the same time.

    Lag.0.1 is our 2x10G between the core 1 and core 2

    This is the Netsight syslog what happens on many stacks:

    false Info --- 05/09/2016 16:01:58 X.X.X.X --- --- Event --- X.X.X.X dot1s_sm.c(3875) 6017223 %% Setting Port(1) Role: ROLE_DESIGNATED | STP Port(1) | Int Cost(2000000) | Ext Cost(2000000)
    false Info --- 05/09/2016 16:01:58 X.X.X.X --- --- Event --- X.X.X.X dot1s_ih.c(1365) 6017222 %% Setting Port(1) instance(0) State: FORWARDING
    false Info --- 05/09/2016 16:01:58 X.X.X.X --- --- Event --- X.X.X.X dot1s_ih.c(1302) 6017221 %% Setting Port(1) instance(0) State: LEARNING
    false Info --- 05/09/2016 16:01:58 X.X.X.X --- --- Event --- X.X.X.X dot1s_ih.c(1423) 6017220 %% Setting Port(1) instance(0) State: DISCARDING
    false Info --- 05/09/2016 16:01:58 X.X.X.X --- --- Event --- X.X.X.X dot1s_txrx.c(450) 6017219 %% dot1sMstpTx(): CIST Role Disabled
    false Info --- 05/09/2016 16:01:58 X.X.X.X --- --- Event --- X.X.X.X dot1s_ih.c(1423) 6017218 %% Setting Port(1) instance(4095) State: DISCARDING
    false Info --- 05/09/2016 16:01:56 X.X.X.X --- --- Event --- X.X.X.X dot1s_ih.c(1423) 6017217 %% Setting Port(1) instance(0) State: DISCARDING
    false Info --- 05/09/2016 16:01:56 X.X.X.X --- --- Event --- X.X.X.X dot1s_txrx.c(450) 6017216 %% dot1sMstpTx(): CIST Role Disabled

    Is there any possibility to find out if a gbic has a malfunction?

    Thanks a lot

    Adam