<div dir="ltr"><div><div><div>I have seen some errors like "gfid different on subvolume" in my deployment.<br>e.g.<br>[2014-03-26 07:56:17.224262] W [afr-common.c:1196:afr_detect_self_heal_by_iatt] 0-sh_ugc4_mams-replicate-1: /operation_1/video/2014/03/26/24/19: gfid different on subvolume<br>
<br></div>my clients (3.3) have already backported the patches mentioned in<br><a href="https://bugzilla.redhat.com/show_bug.cgi?id=907072">https://bugzilla.redhat.com/show_bug.cgi?id=907072</a><br><br>CHANGE: <a href="http://review.gluster.org/4459">http://review.gluster.org/4459</a> (cluster/dht: ignore EEXIST error in mkdir to avoid GFID mismatch) merged in master by Anand Avati<br>
<pre class="" id="comment_text_5"><a href="http://review.gluster.org/5849">CHANGE: http://review.gluster.org/5849</a> (cluster/dht: assign layout onto missing directories too)</pre><br></div>But I still saw such errors.<br>
<br></div><div>I though these changes were relative to clients? am I right? need I update my servers using the patched release?<br><br></div><div>Or I miss other things? <br></div><div><br><br></div><div><br></div></div>