I have a rb532 ver 5.7 with 2 wlans for AP each carries about 15 clients.
wlan1 10.2.2.1/24
wlan2 10.2.3.1/24
These are all good
eth1 has 10.251.250.4/24 gateway 10.251.250.1, received dynamically through a pair of bridged dynamically addressed NBM5s (client 10.251.250.3, ap 10.251.250.2) from a rb493AH 10.251.250.1/24 dhcp server/gateway
eth3 has 192.168.102.2/24 as address usually for admin access and was connected to another 532a (192.168.102.1/24) the old gateway to 0.0.0.0/0
As soon as I allowed the rb532 eth1 to receive dynamic address and routing the old gateway was disabled.
I have routing info to allow access to the board at 192.168.102.2 through the new route.
Everything worked fine for about 24 hours, then the DUDE showed me that 192.168.102.2 was down, but all addresses were coming through, and
I can access it with any port address except 192.168.102.2 ie 10.251.250.4 works and will keep winbox open at least 5 minutes, sometimes kicks me off after that.
Trace from other Routerboards get to 10.251.250.1 then bounce back and forth to 10.251.250.3 (NBM5 Client attached to ethr3)
CPU is running 15-25%
What might be happening here?
Michael
Edits:
OK If I put route table in the NBM5 bridge (client 10.251.250.3) attached to the rb532, everything works.
Is this normal behavior for a bridge?
1st router DHCP/Gateway > Bridge AP >Bridge Client> 2nd router DHCP Client
Router client gets all necessary info from 1st router, but route table must be added to BridgeClient???
Or I am just kludging a fix?
I have not used bridging much, always routed everything…