cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
8747
Views
10
Helpful
5
Replies

Ethernet Giant Frames

Umesh Shetty
Level 1
Level 1

Hi All,

The show interface fax/x output contains a feild called "giants" which according to cisco means "Number of packets that are discarded because they exceed the medium's maximum packet size" My question is the packet size exceeds the maximum packet size will it not be fragmented ? Why does the interface have to drop it, or does this counter only contain packets exceding the size and have the DF bit set ??

Thanks in Advance

Umesh Shett

5 Replies 5

Joseph W. Doherty
Hall of Fame
Hall of Fame

Disclaimer

The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.

Liability Disclaimer

In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.

Posting

Fragmentation happens on L3 egress (unless packets sets the DF bit).

L2 assumes frames, both ingress or egress, don't exceed MTU.

Thnx Joseph,

What if a router receives a 1700 Byte frame ? Wikepedia says that the maximum payload size of an ethernet is 1500 octets (which is also the egress interfaces MTU)  if the routers ingress ethernet interafce starts dropping frames that exceed the 1500 + 18 (ethernet encapsulation) limit there would be no frangemntation ever needed. At no times the ingress interface will ever send an L3 packet which is more that 1500 octets and hence will never exceed the egress interfaces MTU and hence no fragmentation. This is what I am confused about , hope you can help.

Regards

Umesh

Disclaimer

The  Author of this posting offers the information contained within this  posting without consideration and with the reader's understanding that  there's no implied or expressed suitability or fitness for any purpose.  Information provided is for informational purposes only and should not  be construed as rendering professional advice of any kind. Usage of this  posting's information is solely at reader's own risk.

Liability Disclaimer

In  no event shall Author be liable for any damages whatsoever (including,  without limitation, damages for loss of use, data or profit) arising out  of the use or inability to use the posting's information even if Author  has been advised of the possibility of such damage.

Posting

If the receiving (ingress) port was "standard" Ethernet, the received frame would be dropped, as being oversized.

If the receiving (ingress) port was "jumbo" capable Ethernet, jumbos were enabled, and if its jumbos supported 1700 bytes (there's no jumbo size standard), then the frame would be accepted on the port.  What happens next would depend whether the received frame needed to be forwarded (it might be directed to the router itself), and if so, how the egress port was configured.  It might be dropped by an egress port on that router, forwarded as same size, or L3 fragmented into multiple L2 frames.

BTW, fragmentation is at L3, i.e. packets are fragmented, Ethernet frames are L2 and are never fragmented.  Even from the source host, with standard Ethernet end-to-end, packets will be fragmented if they are larger than the frame can contain.  Normally, source host will size its packets to fit into the host's frames without the need to fragment.

Hi Joesh,

Just assuming a network shown below if I were to ping from host A to Host B with 1600 bytes of data. That would be a total of 1600 (data) + 8 bytes (ICMP header) + 20 bytes ( IP Header) + 18 Bytes (Ethernet Header) == 1646 Bytes of ethernet frame. So when R1's ingress interface receives the frame and decapsulates it , it must have an L3 packet of 1628 bytes and If I do not set the df bit in the ICMP packet i would expect the packet to be fragmented on R1's egress interfaces(connected to R2). 

Thanks in Advance

Umesh

Disclaimer

The   Author of this posting offers the information contained within this   posting without consideration and with the reader's understanding that   there's no implied or expressed suitability or fitness for any purpose.   Information provided is for informational purposes only and should not   be construed as rendering professional advice of any kind. Usage of  this  posting's information is solely at reader's own risk.

Liability Disclaimer

In   no event shall Author be liable for any damages whatsoever (including,   without limitation, damages for loss of use, data or profit) arising  out  of the use or inability to use the posting's information even if  Author  has been advised of the possibility of such damage.

Posting

For what you described to work as described, would require host 1 (in you drawing or host A in your text) and R1's interface to that host to support jumbo Ethernet and the rest of the interfaces standard Ethernet.

Review Cisco Networking products for a $25 gift card