We have a new UTM 120 as a replacement for a Cisco ASA.
Setup is as follows:
2 LANs (DMZs)
1 WAN Connection
LAN Interfaces are connected to two Cisco Catalyst Gigabit Switches.
Now we have problems when Server1 in LAN1 makes larger SQL queries (~ 100MB output) to Server2 in LAN2 through the UTM 120. Sometimes the query is working but sometimes the query simply hangs and the connection seems to be totally lost (after ~16 MB).
I tried changing cables, switch ports, configuring ports and NICs from Gbit to 100 MBit FD and also downgraded from 9.1 to 9.0. => same results.
We noticed dropped RX packets in output from ifconfig on both LAN interfaces on the UTM. Number increases when we start these bigger queries.
I tested the throughput with iperf between the two servers. Best result is ~450 MBit/s but number of dropped RX packets increases while iperf is testing.
Any ideas where the dropped packets come from? I am at a loss now. [:(]
This thread was automatically locked due to age.