Date: Fri, 15 Oct 2021 22:43:41 UTC
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=259152 --- Comment #8 from John Baldwin <jhb@FreeBSD.org> --- Created attachment 228734 --> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=228734&action=edit permit_expdatasn_0.patch Not tested, but I did notice that we should require an ExpDataSN of 0 if the response is not command completed. Looking at the latest trace, it does seem that the SCSI response does include an ExpDataSN of 0, and I've added an exception that tries to exempt this check for errors. It's not conformant, but Linux also doesn't check for this that I can see (linux/drivers/scsi/libiscsi.c:iscsi_scsi_cmd_rsp()). This though will just get you back to comment #4 I think and we will still need a pcap with -s 0 of the result to look at for the next issue. -- You are receiving this mail because: You are the assignee for the bug.