glusterfs-3.8.4-54.15.el7

Asianux Gluster Storage is a software only scale-out storage solution that
provides flexible and affordable unstructured data storage. It unifies data
storage and infrastructure, increases performance, and improves
availability and manageability to meet enterprise-level storage challenges.

This advisory fixes the following bugs:

* Previously, glusterd could not handle blank real paths when checking if brick path is already a part of another volume, during volume create. Hence, volume create requests would fail with the error 'Brick may be containing or be contained by an existing brick'. This update fixes the path comparison logic to correctly handle blank paths, and thus prevents failure of further volume create requests. (BZ#1599803)

* Previously, glusterd could not check if the daemons it initiated, were fully initialized before sending them requests. Hence, if glusterd sends the index heal request from CLI to the self-heal daemon before it fully initializes its graph, the self-heal daemon would crash. This update fixes the self-heal daemon by ignoring the requests it receives from glusterd before graph is initialized. Thus, the CLI fails the command when user launches the index heal via. gluster CLI. (BZ#1595752)

* Previously, eager-lock was disabled for volumes hosted by a block, because conflicting writes were handled incorrectly when eager-lock is enabled. Hence, the performance of gluster backed block devices was insufficient when eager-lock was enabled. This update fixes the eager-lock handling for conflicting writes. Thus, when eager-lock is enabled, performance of gluster backed block device is enhanced. To observe this performance improvement, the Gluster administrator needs to enable eager-lock on old block hosting volumes. Also, the eager-lock option is enabled by default for all new volumes. (BZ#1583733)

* Previously, due to a bug in glusterd, executing the `gluster volume set client-io-threads on` command on a replicate volume returned success without actually enabling it in the client graph. This updates fixes that bug, and executing the `gluster volume set client-io-threads on` command provides a correct success result and ensures that the translator is loaded on the client graph. (BZ#1598416)

* Previously, when an application sent fsync on a sharded file, the shards associated with the file were not being synced to disk. Hence, potentially causing data loss in case of a plain distribute shard volume. This update fixes the shard translator by syncing all the modified shards to disk whenever the application sends an fsync. (BZ#1583462)

All users of Asianux Gluster Storage are advised to upgrade to these updated packages, which resolve these issues.