sivanov and andersen , I looked the data from both of you and determined that the problem is not with the emlxs driver. The problem is an incompatibility between your Brocade switch firmware and COMSTAR. When COMSTAR issues CT requests to the switch it is now setting the CT Revision level in the request to 2. I believe it used to use revision level 1. These requests are being rejected by the switch with a reason code which indicates an "Invalid version level" error. This means that the Brocade switch can't handle revision level 2 requests from COMSTAR. I looked into COMSTAR source code and found that if a switch rejects the GID_PN request (CT cmd 0x0121) it will automatically drop the connection from the initiator port the command is referencing. That is why the initiator suddenly can’t see the target anymore. The slight differences in behavior between build 122 & 130 is probably slight differences in the COMSTAR code and timing. It was mentioned that some connections were "flakey". That usually indicates a timing related problem. But ultimately all the issues appear to go back to the Brocade firmware and how COMSTAR is reacting to the rejected requests.
I would suggest contacting Brocade to see if they have newer firmware for their switch's that support CT revision level 2 requests. I don't know if they do or not. -- This message posted from opensolaris.org _______________________________________________ storage-discuss mailing list storage-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/storage-discuss
