Subject:
How to check if 9000 MTU/Jumbo Frames are working.
MTU missmatches can be the cause for dropped and overrun packets.
Confirm that your switches and clients also are all configured for the same MTU value.
Switch port configuration problems can cause dropped packets as these are dropped packets on the RX incoming to the QuantaStor, the source of the Dropped packets would be external.
Please investigate the MTU of the clients and switch port logs to confirm the source of the errors.
Details:
Run this command to check for Jumbo Frames:
ping -M do -s 8972 [destIP]
Note: The reason for the 8972 on Linux devices is that the ICMP/ping implementation doesn’t encapsulate the 28 byte ICMP - "(8) + IP (20) (ping + standard internet protocol packet)" header – so we must take the 9000 and subtract 28 = 8972.
If everything is setup correctly you should see an output like this:
PING 192.168.65.195 (192.168.65.195) from 192.168.65.195 eth1: 8972(9000) bytes of data.
8980 bytes from 192.168.65.195: icmp_req=1 ttl=64 time=0.372 ms
8980 bytes from 192.168.65.195: icmp_req=2 ttl=64 time=0.066 ms
If you have not enabled jumbo frames/9k MTU on your client device you’re sending the ping from you’ll see:
PING xxx.xxx.xxx.xxx (xxx.xxx.xxx.xxx): 8184 data bytes
ping: sendto: Message too long
If you have jumbo frames enabled on your client but not the destination or on a switch in between you will see an output like this:
PING xxx.xxx.xxx.xxx (xxx.xxx.xxx.xxx): 8184 data bytes
Request timeout for icmp_seq 0
Please review this link for additional information on QuantaStor network configuration:
QuantaStor_Administrators_Guide#Network_Port_Configuration
Comments