3

Recently we've configured 4 HP switches which are using Distributed trunking. See picture

enter image description here

So the problem is this.

Rack 4 has 2 switches (Left on the picture). One of these switches is the root bridge for the STP instance.

Rack 6 has also 2 switches (Right on the picture). One of these switches is the root bridge for this STP instance.

What is failing:

From 10.0.0.13 I am unable to ping 10.0.0.17 and vice versa. I am however able to ping the directly connected switches from 10.0.0.13 as well as from 10.0.0.17

What is going on?

EDIT 1: Trunk status

r11-06-sw-10# show trunks 

 Load Balancing Method:  L3-based (default)

  Port   | Name                             Type       | Group  Type    
  ------ + -------------------------------- ---------- + ------ --------
  A8     | Trk99-ISC                        SFP+DA1    | Trk99  LACP    
  B7     | Trk1-r11-05-sw-11                SFP+DA3    | Trk1   dt-trunk
  B8     | Trk99-ISC                        SFP+DA1    | Trk99  LACP    
  E24    | Trk10-duocast                    100/1000T  | Trk10  dt-lacp 
  F8     | Trk20-r11-04-sw-10               SFP+DA3    | Trk20  dt-lacp 

Show lacp peer

r11-06-sw-10# show lacp peer 

LACP Peer Information.


System ID: 3464a9-a10300 


  Local  Local                       Port      Oper    LACP     Tx   
  Port   Trunk  System ID      Port  Priority  Key     Mode     Timer
  ------ ------ -------------- ----- --------- ------- -------- -----
  A8     Trk99  3464a9-9fa300  8     0         1060    Active   Slow 
  B8     Trk99  3464a9-9fa300  40    0         1060    Active   Slow 
  E24    Trk10  00239c-16b600  52    127       53      Active   Fast 
  F8     Trk20  00fd45-5eac00  10161 0         981     Active   Slow 

Trunk status:

r11-04-sw-11# show trunks 

 Load Balancing Method:  L3-based (default)

  Port   | Name                             Type       | Group  Type    
  ------ + -------------------------------- ---------- + ------ --------
  A7     | Trk30-westpoort-sw-10            SFP+LR     | Trk30  dt-lacp 
  A8     | Trk99-ISC                        SFP+DA1    | Trk99  LACP    
  B6     | Trk10-r11-04-jnp-10              SFP+SR     | Trk10  dt-lacp 
  B7     | Trk20-r11-06-sw-11               SFP+DA3    | Trk20  dt-lacp 
  B8     | Trk99-ISC                        SFP+DA1    | Trk99  LACP   

Show lacp peer:

r11-04-sw-11# show lacp peer 

LACP Peer Information.


System ID: 00fd45-618100 


  Local  Local                       Port      Oper    LACP     Tx   
  Port   Trunk  System ID      Port  Priority  Key     Mode     Timer
  ------ ------ -------------- ----- --------- ------- -------- -----
  A7     Trk30  1458d0-f50400  40    0         415     Active   Slow 
  A8     Trk99  00fd45-5eac00  8     0         1060    Active   Slow 
  B6     Trk10  407183-5e7380  2     127       1       Active   Fast 
  B7     Trk20  3464a9-9fa300  10161 0         981     Active   Slow 
  B8     Trk99  00fd45-5eac00  40    0         1060    Active   Slow 
7
  • 1
    You need to edit your question to include the switch configurations, interface status and statistics, logs, etc. Any one VLAN is only going to have one root bridge, so if the switch addresses are all in the same network and VLAN, then there is only one root bridge. If you have multiple VLANs, then you need a router to get from one to the other. – Ron Maupin Oct 15 '17 at 14:45
  • What type of switches or those? Distributed aggregation is not supported by all HP models. What's the status of the trunks on the switches? LACP or static? – Zac67 Oct 15 '17 at 14:46
  • See update. These switches are doing Layer 2 switching. Only management vlan has an IP so I can connect to it. – Valentin Bajrami Oct 15 '17 at 14:53
  • 1
    Again, please include the configurations for all the switches. We cannot simply guess what is wrong, and the question is too broad without all the information. – Ron Maupin Oct 15 '17 at 15:01
  • 1
    There can only be one root per VLAN. If you cannot provide the information we need to help you, then we cannot help you. You can obfuscate any public addresses or passwords, but simply speculating or guessing is off-topic here. This is not a discussion forum, we have Network Engineering Chat for that. – Ron Maupin Oct 15 '17 at 15:18
2

Trunk the a port on each pair dt-switch 1 (.12 and .13) then do the same in dt-switch 2 (.17 and .18). This trunk should be physical connected to each facing dt-switch pair or cross connection.

This will give the idea to each device that each facing dt-switch are one thus able to ping each network subnet as long proper tag or member of the VLANS you have

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.