Hola.   Consulta. La lentitud afecta a todas las vlans que pasan por el vínculo o solo algunas?   podrías compartirme lo siguiente de ambos equipos?   En ambos:   sh run show spanning - tree detail   show spanning - tree detail  |  inc ieee |occurr|from|is  exe show process cpu sorted | e 0.00 show ip traffic   En SW_CORE_1:   show platform qos queue stats gigabitEthernet 1/0/19 show platform qos queue config gigabitEthernet 1/0/19   En SWCORE_EDIF2:   show platform qos queue stats gigabitEthernet 1/0/24 show platform qos queue config gigabitEthernet 1/0/24     Saludos ... View more

Hola   Podrías hacer la siguiente prueba.   default de la interface aplicar los cambios. grabar la conf verificar con un show startup que la nueva config esta aplicada para el puerto desconectar y reconectar el puerto   NOTA: Si estas utilizando tacacs, verificar adicionalmente si en el syslog de tacacs no aparecen mensajes al momento de la tarea.   Saludos ... View more

Hola Proesco.   Estimo que los switches son de capa 2 y hay uno de ellos que es de capa 3. Suponiendo todo este escenario. Los equipos de capa 2 al hacer el ping estarán saliendo con source del tráfico con su ip de gestión y a menos que la misma este dentro del rango 51.x el mismo deberá ir primero al DG para recién ahí ser ruteado al destino. Esto ocurre, aun cuando vemos ambos equipos conectados directamente.  Suponiendo que el equipo de capa 3 o DG de ambas vlans es el sw 3560 con ip 51.254. Deberías validar que el mismo tiene configurado el comando "ip routing" para poder hacer el ruteo intervlan. Esto es por que aunque ambos switches (51.213 <--> 51.211) figuren directamente conectados, el tráfico del ping estaría llendo primero al DG (51.254) para recién volver al host. Esto nos dice también, que en caso de que el vínculo de las antenas cuente con algun fiiltro o ACL, tambien esto podría sumar el problema que tienes.       Saludos ... View more

Hola José.   Contarías con lo siguiente?   sh run show spanning - tree detail   show spanning - tree detail  |  inc ieee |occurr|from|is  exe show process cpu sorted | e 0.00   Saludos ... View more

Hi   I share an interesting link regardin MTU on 37xx devices. You can try to enable jumbo frames. Also for traffic between hosts in vlan 25 to hosts in vlan 20 you can test the MTU that Ubiquitys are configured for.   // Cisco Catalyst 3750/3560 Series Switches - Jumbo/Giant Frame Support Configuration https://support.hpe.com/hpsc/doc/public/display?docId=c03331307   BR Gaston ... View more

Hi   Can you share a show version?   BR Gaston ... View more

Hi   Do you have a policy-map configured on the device?   Can you try to enable a policy-map on an interface? this example should enable the use of the soft-max configuration you mentioned earlier.   // Example conf t policy-map test exit   interface gx/y  service-policy output test end   Then verify with the commands to see the queues.   Additionally, can you share a show run?   I also share an interesting link related to output drops on 3850   https://www.cisco.com/c/en/us/support/docs/switches/catalyst-3850-series-switches/200594-Catalyst-3850-Troubleshooting-Output-dr.html   BR Gaston ... View more

Hi.   Some questions.   Does this device boots completely? or it continues booting in a loop? Does this device has some modules? please try removing them and then try again.   BR Gaston       ... View more

Hi   Option 1:   Please try to ping from 4k5   172.31.0.3 with source 172.31.0.5   If this does not work, do correct the routing (via static or dynamic)   Option 2: (more common)   source the tunnel interface from phisical interfaces instead of loopbacks   BR Gaston ... View more

Hi.   Some questions.   Can you ping each of the other sides of the Point-to-Point interfaces? From HQSC-IEG-1 ping vrf HQSC-MTNEDGE-VRF 10.20.0.2 ping vrf HQSC-MTNEDGE-VRF 10.0.0.x   for subinterfaces 0/0/0.12 and 0/0/1.12 are the other sides in a vrf also? or the other sides are in a global routing table? I ask you this, because, it seems that may be devices in network 10.20 does not have the way back to network 10.0.0.1 and vice versa.   this can be solved adding a default route or a DG on those devices pointing to HQSC-IEG-1. If the interfaces on those devices are inside a vrf, the default route or DG must be in that vrf. If the interfaces are in the global routing table, with just a static route / DR or DG should fix the problem.   Examples:   From device connected to 0/0/0.12 ip route 0.0.0.0 0.0.0.0 10.0.0.1 or (in case the other side be in the same vrf) ip route vrf HQSC-MTNEDGE-VRF 0.0.0.0 0.0.0.0 10.0.0.1     From device connected to 0/0/1.12   ip route 0.0.0.0 0.0.0.0 10.20.0.1 or (in case the other side be in the same vrf) ip route vrf HQSC-MTNEDGE-VRF 0.0.0.0 0.0.0.0 10.20.0.1   BR Gaston   ... View more

Hi.   This behavior that you are seeing in your production network. it happens to all vlans? I mean, several vlans with out a DG can ping to others vlans? or it just happens for 2 vlans?   If the issue affects only 2 vlans, please check in your trunks that you may not be doing a native vlan missmatch. Because with it you can end up joining two separete bcasts domains.   BR. Gaston ... View more

Hi   Please do the next procedure.   Remove all the linecards from the box except supervisors. Then, power on the box. It ought to be enough to let the complete boot up process finilize   If this work ok, then please check your power supplies to certify that you have power enough to boot up all the linecards at the same time.   BR Gaston ... View more