Hello guys
I'd want to submit a very trying problem for me. This is my network config:
- core switches: pair of nexus 5548up in vpc domain with HSRPv2, 40 vlans, 1 SVI per vlan, 1 HSRP group per vlan, downlink to dist sw with LACP and port-channel, RSTP enabled, PIM sparse mode enable
- dist switches: stack of catalyst 3750x-24s with dhcp server enabled per each vlan , downlink to access sw with LACP
- access switches: catalyst 2960s with up to 48 mediaplayers clients (vlan from 11 to 50)
- servers attached to core switches (vlan 10)
all works fine, until physical links between cores and dist sw are up, when they are cutted by a dedicated interface and then they go up again, all clients are able to get IP address from dist sw, but in some vlans (randomly) some of clients (within the same vlan) are not able to reach servers.
Troubleshooting:
- vPC formed and consistency ok, peer-gateway and peer-swicth enabled
- STP ok, all ports in all vlans are in forwarding state
- hsrp active/standby ok
- LACP links ok
- unicast routing ok for working clients
- multicast routing ok
- L2 table in both cores ok
- L3 table in both cores ok
- from servers I cannot ping some clients
to fix the problem I have to shutdown and no shutdown the SVI in the primary core related to vlans with not working clients, but I need to fix permanently this behaviour. Did anybody have similar problem or could anybody guess what's the issue?
Thank you in advance.
Marco
