2012/11/27 Steve French <smfrench@xxxxxxxxx>: > Am curious how you noticed this problem? Did you have a test case? The test case that reproduces the problem: 1) client1 opens "filename" (file that already has enough data ~~ 4096 bytes is ok) 2) client1 sets a read lock a file from pos=1 to pos=2 3) client2 opens "filename" - client1 gets oplock break to level II and push lock from #2 to the server. 4) client1 writes 'a' to a file (from pos 0) - got status OK, but the server reject the write (due to the client1 writes the whole page from pos=0 to pos=4096 and conflicts with the read lock from pos=1 to pos=2) - ERROR. 5) client2 reads 1 byte and got the old data (not 'a') because the server returns it's version of the data It is suitable for both CIFS and SMB2. -- Best regards, Pavel Shilovsky. -- To unsubscribe from this list: send the line "unsubscribe linux-cifs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
