5.6.4 HP StoreAll Storage Release Notes (AW549-96075, June 2013)

• The CIFS/SMB service failed when a large number of CIFS/SMB clients were connected
concurrently for read/writes.
• The NIC monitor failed with a segmentation fault.
• Servers becoming unresponsive during connection/delegation recovery. The problem occurred
during connection/delegation recovery on servers especially within a large StoreAll cluster.
• An lwiod failure caused the CIFS/SMB service to stop. This issue has been resolved.
• If the quotamonitor service could not access a server, the service issued an error and did not
obtain other quota information.
• The "ibrix_auth -t -S <sharename> ..." command was failing with the output
(LW_ERROR_DATA_ERROR). The cached data is incorrect whenever it encountered
an unprovisioned user in Active Directory.
Users can run the "ibrix_auth -t -S <sharename> ..." command successfully.
• When segments reached 70–80% of capacity, clients were unable to write to a CIFS share.
• The quota current size would be reduced after running the rebalance task, even though files were
not modified or deleted. During movement of files across segments, there are intermediate replica
files that are created. Once data from an original file is synced with its replica, the original file
is deleted and replica is promoted as the original file. These replicas have to be marked with the
directory quota ID same as that of the original master file, but that was not occurring. As the
original file is deleted post replication, the directory quota accounted for it gets decremented.
Since the newly created file was not marked with the directory quota ID, it was never accounted
for. As a result, a reduction in quota usage was shown post replication.
The fix was to mark the replica created with the quota ID same as that of the original file, so any
data written into this file is accounted for against the set quota ID.
• Data tiering could result in zero length files.
• CIFS/SMB shares were inaccessible, and the Windows client accessing the filesystem using
CIFS/SMB would be shown the message: The specified server cannot perform the
requested operation. The cause was an SMB 2.0 credit exhaustion issue that was seen on
applying a specific load pattern for a prolonged period of time on an HP-SMB server. The fix
addresses the credit leak that was a result of corruption in the inherent data structure
implementation.
• There were too many clients registering for DCN (Directory Change Notification) and the
notification queue would get overloaded, leading to an ASSERT resulting in a segmentation fault
of the Lwiod daemon. The fix adds proper handling of the overload situation, thereby avoiding
the ASSERT.
• During a mkdir operation, the xdr_encode_netobj dereferencing NULL
xdr_netobj->data pointer message was displayed. The underlying issue has been resolved
so the message no longer appears.
• The root file system was constantly running out of space. This release provides the option, -l, for
placing log files in an alternative location for the ibrix_fsck -p 1 and ibrix_fsck -p 2
commands. Use the -l option to prevent the root file system from running out of space. For
example:
ibrix_fsck -p 1 -l /home/user/
ibrix_fsck -p 2 -l /home/user/
In these examples, /home/user/ is the alternative location.
6 5.6.3 release