<div dir="ltr"><div><div><div><div><div>We use XFS on Amazon EBS on Ubuntu 12.04.3, kernel "3.2.0-54-virtual". <br><br></div>Our backup scripts invoke "xfs_freeze -f" before taking a snapshot to ensure filesystem consistency. After the snapshot is taken, we invoke "xfs_freeze -u".<br>

<br>Recently, we have observed that some of these unfreeze calls exit with code 255. This results in a frozen filesystem, processes queuing up and us having to reboot the EC2 instance. Invoking freeze -u manually results in multiple processes like this - <br>

<br>
                                                         17628 ?       
                                                    D      0:00
                                                    /usr/sbin/xfs_io -F
                                                    -r -p xfs_freeze -x
                                                    -c thaw
                                                    /mysql-vol-a<br>
                                                         17694 ?       
                                                    D      0:00
                                                    /usr/sbin/xfs_io -F
                                                    -r -p xfs_freeze -x
                                                    -c thaw
                                                    /mysql-vol-a<br>
                                                    <br></div>A search on this list reveals at least one other instance of this problem - <a href="http://oss.sgi.com/archives/xfs/2013-01/msg00437.html">http://oss.sgi.com/archives/xfs/2013-01/msg00437.html</a> <br>

</div><br>No resolution seems to have happened there. Is this a known bug, or has it been fixed since (I'm running xfsprogs 3.1.7) ?<br><br></div>Regards<br></div>Hrish<br></div>