F80c - Wan1 pppoe - Wan2 MPLS - Internal with HP 2530 58G

[kino_ed]

Hi everybody

I have a Fortigate 80C connected to Wan1 a pppoe services and Wan2 to MPLS with Router cisco, and the FW is in mode NAT,

I have 5 VLANs managed by the FW, with a routing table for to reach the intranet sites through MPLS, apparently all works without problem, internet, ping, tracert, nslookup, all services, the navigation in internet is OK, but the problem that I have is when I try to reach the page through the MPLS take a long time to load it 5 or 10 minutes, but I have detected that this only occur when the devices are connected to 1GB, because the devices that have network card to 100MB or use the Wifi to 130 MB works very well, in fact I have modified the speed in the port of the Switch to make a test and works without problem.

I have verified the interface in the FW, make a sniffer, but is not clear for me because only I have this issue with 1GB speed.

Somevary else know if I need to make some changes in the FW to avoid this problem.

As you can see in the diagnostic of the WAN2 don't see any problem

FGT80C # diagnose hardware deviceinfo nic wan2

Driver_Name                     iegbe

Driver_Version                  0.8.0-NAPI

PCI_Vendor                      0x8086

PCI_Device_ID                   0x5044

PCI_Subsystem_Vendor            0x0000

PCI_Subsystem_ID                0x0000

PCI_Address                     5:1.0

PCI_Bus_Type                    PCI Bus #05

PHY_Type                        3

MAC_Type                        8

IRQ                             0

System_Device_Name              wan2

Current_HWaddr                  00:09:0f:ef:bf:27

Permanent_HWaddr                00:09:0f:ef:bf:27

Link                            up

Speed                           1000

Duplex                          full

State                           up(0x1203)

MTU_Size                        1500

Rx_Packets                      117092

Tx_Packets                      123510

Rx_Bytes                        96670497

Tx_Bytes                        15232733

Rx_Errors                       0

Tx_Errors                       0

Rx_Dropped                      0

Tx_Dropped                      0

Multicast                       398

Collisions                      0

Rx_Length_Errors                0

Rx_Over_Errors          0

Rx_CRC_Errors                   0

Rx_Frame_Errors                 0

Rx_FIFO_Errors                  0

Rx_Missed_Errors                0

Tx_Aborted_Errors               0

Tx_Carrier_Errors               0

Tx_FIFO_Errors                  0

Tx_Heartbeat_Errors                     0

Tx_Window_Errors                0

Thanks a lot for your help

[ede_pfau]

As far as I can guess there are 2 reasons:

1- the cable between WAN2 and the Cisco router is broken

2- the Cisco router port doesn't negotiate well, either speed or duplex or polarity

For 1-, change the cable to a proven new one. For FE only 4 of the 8 wires are used, for GE you need all 8.

For 2- temporarily put a 100/1000 switch in between which is capable of buffering traffic.

Maybe these simple tests will give you more hints where to locate the source problem. I don't think MTU is an issue here as it would have been reflected in the port error statistics. BTW, you should check the error stats of the Cisco port as well.