Dr. Volker Jaenisch wrote: > Runs without problems over TCP over IB. We have this setting in production since a while. > Please have a look on my recent postings in open-iscsi mailing list. [...] > Yes, but I also tried on ConnectX HCAs wich reproduces the read errors perfectly. OK, this morning I did some reading over your previous postings to open-iscsi and stgt and also some testing. The news is that I have this reproduced on my setup. When I use a different target, it doesn't happen! so I think we're quite narrowed this to be stgt's fault, we'll see... Is it correct that at some point you said that a patch of Pete which deferrers sending the scsi response till the rdma-write is completed makes the problem disappear? do you have this patch in a form that applies on tgt's head? Also, just to make sure we're on the same page: for write you use opat=1 and for read ipat=1 mismatch=1, correct? Or. -- To unsubscribe from this list: send the line "unsubscribe stgt" in the body of a message to majordomo at vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html |