I'm currently running 2 ceph clusters (ver. 0.80.1) which are providing secondary storage for CloudPlatform. Each cluster resides in a different datacenter and our federated gateway consists of a region (us-east-1) with 2 zones (zone-a [master], zone-b [slave]). Objects appear to be replicating/syncing from zone-a to zone-b as expected, meaning the objects appear in both zones with the same size and checksum using an s3 client to view. We've recently run into an issue where an object is replicated to zone-b, the object appears to be complete, yet the .us-east-1.zone-b.rgw.buckets pool continues to fill with shadow files for this object. We noticed the osd's were being consumed rather quickly, and while troubleshooting we found 230+ unique TAGS for the object (i.e. TAG_shadow_1 through TAG_shadow_515). Has anyone seen this behavior or have any idea what may have caused it. Thanks in advance for any help that may be provided, MLM -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.ceph.com/pipermail/ceph-users-ceph.com/attachments/20140919/0202922e/attachment.htm>
 |