Hi Glusterfs Users!<br><br>I have got one replicated volume with two bricks:<br><br>s1 ~ # gluster volume info<br><br>Volume Name: data-ns<br>Type: Replicate<br>Status: Started<br>Number of Bricks: 2<br>Transport-type: tcp<br>
Bricks:<br>Brick1: s1:/mnt/gluster/data-ns<br>Brick2: s2:/mnt/gluster/data-ns<br>Options Reconfigured:<br>performance.cache-refresh-timeout: 1<br>performance.io-thread-count: 32<br>auth.allow: 10.*<br>performance.cache-size: 1073741824<br>
<br><br>There are 5 clients which have got mounted volume from s1 server. <br><br>We've face a hardware failure on s2 box for about one week. During that time the s2 box was down.<br>All read writes operations went to s1.<br>
Now I would like to synchronize all files on s2 which is operable. I have started Glusterfs Server and <br>executed self healing process("find with stat"on the glusterfs mount from s2 box). <br>During the replication process I have faced very strange behaviour of Glusterfs.<br>
Some of clients have tried to get lots of files from s2 server, but those files did not exist or have got 0 bytes size. <br><br>It caused lots of "disk wait" on the web servers (clients which have got mounted volume from s1) and finally 503 http response had been sent.<br>
<br>My question is, how to avoid serving files from s2 box until all files would be replicated correctly from s1 server?<br><br>I have installed Glusters 3.2.6-1 from Debian repository.<br><br>Thank you a lot in advance, <br>
Jimmy,<br><br>