On 19 Sep 2016, at 10:58, Christoph Hellwig wrote: > So it looks like this patch go in, but when I did my usually NFS > regression tests that I stopped for a few weeks aover the summer it > broke down badly: all the fsx and some other data integrity tests now > fail in xfstests over pNFS blocklayout. > > Given how close it is to 4.8-final can we revert it for now? Are you sure this patch is the problem? I just tested and things are broken for me too, but I think it has something to do with the client choosing the block layout over SCSI now that the server is returning multiple layout types. I don't have block layout configured properly, but still I don't think I should be getting -EIO if it isn't configured. Shouldn't IO fall back to the MDS? It works if the server only returns SCSI layout type.. unless you're testing something I'm not.. Just wondering if you reverted this patch and saw the problem go away. Ben -- To unsubscribe from this list: send the line "unsubscribe linux-nfs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
