<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=us-ascii">
<meta name="Generator" content="Microsoft Exchange Server">
<!-- converted from rtf -->
<style><!-- .EmailQuote { margin-left: 1pt; padding-left: 4pt; border-left: #800000 2px solid; } --></style>
</head>
<body>
<font face="Calibri" size="2"><span style="font-size:11pt;">
<div>Description of problem:</div>
<div>My setup has 5 gluster volumes, and each of them has 2 bricks as backend. </div>
<div>When I copy a large file (100MB) in a gluster volume, 9/10 times it works OK. But about 1 in 10 times the resulting md5 is wrong. After checking I found that the file in one brick has the correct md5sum, while the file in the other brick has a wrong md5sum.
The size of the two files is the same.</div>
<div>By running "cmp -l <correct_file> <wrong_file>"</div>
<div>I found that the difference was in 49 bytes. So the files in the two bricks had the same size, but 49 files were different. Interestingly enough I saw the same number of 49 bytes being different at every check that I made.</div>
<div> </div>
<div>Do you know what might cause this behavior, has anyone seen something like this before? Is this a bug in glusterfs? </div>
<div> </div>
<div> </div>
<div>Version-Release number of selected component (if applicable):</div>
<div>glusterfs 3.7.5 built on Nov 19 2015 16:29:59</div>
<div>Repository revision: git://git.gluster.com/glusterfs.git</div>
<div>Copyright (c) 2006-2011 Gluster Inc. <<a href="http://www.gluster.com">http://www.gluster.com</a>></div>
<div>GlusterFS comes with ABSOLUTELY NO WARRANTY.</div>
<div>You may redistribute copies of GlusterFS under the terms of the GNU General Public License.</div>
<div> </div>
<div> </div>
<div>How reproducible:</div>
<div>Not easy to reproduce, about 1 in 10 times in some environments, not reproducible at all in other environments.</div>
<div> </div>
<div>Steps to Reproduce:</div>
<div>1. scp <100MB file> <path in gluster volume></div>
<div> </div>
<div>Actual results:</div>
<div>1. md5sum of destination should be the same as the source</div>
<div>2. If checksum of files is different between the two bricks, the command "gluster volume heal <vol-name> info split-brain" should return that the two bricks are in split-brain.</div>
<div> </div>
<div> </div>
<div>Expected results:</div>
<div>1. 1 in 10 times the destination file has incorrect checksum. Size is the same, but 49 bytes are altered.</div>
<div>2. "gluster volume heal <vol-name> info split-brain" does not return that the bricks are in split-brain, even though the checksum of the file in the two bricks is different. The size of the file is the same in the two bricks. But 49 bytes are altered.</div>
<div> </div>
<div>Additional info:</div>
<div> </div>
</span></font>
</body>
</html>