[Open-FCoE] Problem with large operations

Mikkel Hagen mhagen at iol.unh.edu
Fri Apr 25 21:22:48 UTC 2008


After further investigation it looks the ELS_REC occurs in an error path 
on the initiator when it does not receive a response from the target 
within 10sec.  So I see two issues here:
1. The target stops responding to large read/write requests.
2. The target does not support ELS_REC.

Is anyone else seeing this or are they known issues? Thanks!

Mikkel Hagen
Project Assistant - Fibre Channel/SAS/SATA Consortiums
Research and Development Engineer - iWARP Consortium	
FC/SAS/SATA:1-603-862-0701  iWARP:1-603-862-5083  Fax:1-603-862-4181
UNH-IOL
121 Technology Drive, Suite 2
Durham, NH 03824



Mikkel Hagen wrote:
> The following command works well:
>  dd if=/dev/zero count=1 bs=512 of=/dev/sdb
>
> When I try and execute a larger operation:
>  dd if=/dev/zero count=1000 bs=512 of=/dev/sdb
>
> I am seeing the following in wireshark:
> 0000   fc fc fc 01 01 02 fc fc fc 01 01 01 89 06 00 00
> 0010   00 00 00 00 00 00 00 00 00 00 00 2e 22 01 01 02
> 0020   00 01 01 01 01 29 00 00 00 00 00 00 02 cb ff ff
> 0030   00 00 00 00 13 00 00 00 00 01 01 01 02 c7 ff ff
> 0040   f5 a1 1a d1 42 00 00 00
>
> Which is decoded as a FC ELS REC command.
>
> The response is:
> 0000   fc fc fc 01 01 01 fc fc fc 01 01 02 89 06 00 00
> 0010   00 00 00 00 00 00 00 00 00 00 00 2e 23 01 01 01
> 0020   00 01 01 02 01 98 00 00 00 00 00 00 03 26 02 1e
> 0030   00 00 00 00 01 00 00 00 00 0b 00 00 bd 84 2d c5
> 0040   42 00 00 00
>
> Which is decoded as an LS_RJT reason code command not supported. After 
> this exchange I am seeing ABTS and the initiator and target seem to lose 
> their connection and need to log back in and re-scan the luns. The dd 
> operation works on count=1, 10 and 100 it is when I go to 1000 when it 
> seems to break.
> Any ideas? Thanks!
>
>   



More information about the devel mailing list