3.7.0 HP StorageWorks HP Scalable NAS File Serving Software release notes for Linux (AG513-96006, October 2009)
DescriptionDefect
Nodes stall waiting for locks
If you are seeing alerts stating that nodes are stalled waiting for locks for a particular
filesystem, the filesystem may be experiencing contention on Full Zone Bitmaps. To aid in
diagnosing this problem, determine whether the following apply:
• Full Zone Bitmaps (FZBMs) are enabled on the filesystem.
• Filesystem operations that involve allocating or freeing blocks are slower than on previous
versions of the filesystem.
• The workload on the filesystem is doing allocations or frees from multiple nodes at once.
To determine whether FZBMs are appropriate for your filesystem, review the guidelines in
the article Using the FZBM On-Disk Filesystem Format on www.hp.com.
9615
HP Scalable NAS cannot detect all imported disks
If HP Scalable NAS is failing to detect and configure disks that you know to be physically
present, you may be encountering a known problem in the Linux operating system.
To work around the problem, adjust the max_lun parameter in the /etc/opt/hpcfs/conf/
scl.conf file. Locate the following lines in the file, and set the parameter to the highest
LUN number.
# max_lun
# In the event that the storage doesn't properly report
# its visible LUNs, AND if the LUNs are not numbered
# contiguously, this parameter can be modified to force
# SCL to probe for non-contiguous LUNs. The value should
# be set to the highest visible LUN id.
# Normal case is this parameter should not be needed, even
# if LUNs are non-contiguous, because most storage can
# correctly report its LUNs, in which case SCL will use
# that data).
#
#max_lun 16 # modify this value to the actual highest LUN
10113
Using hostnames in .matrixrc file can cause connection delays
When servers are specified by hostname in the .matrixrc file, long connection delays
(possibly minutes per hostname entry) can occur if there is a slow or unresponsive DNS server
on the network. During this time, the Management Console and mx commands might be
unresponsive.
10191
Simultaneous 16-node SNMP transactions to Brocade switches can fail
Problems have been seen when a large number of servers (16) simultaneously attempt to use
SNMP to communicate with Brocade switches. When this happens, the switch does not
respond in a timely manner to the SNMP requests. This problem is primarily seen when
exporting a configuration to all 16 nodes at once via mxconfig.
Workaround. Export the configuration to smaller groups of nodes. For example, the operation
will succeed if you export to just four nodes at a time. Intermittent alerts reporting that the
switches are not responding may also appear when multiple servers are being started
simultaneously. This is a transitory condition and should cease when all servers are started.
10762
Management Console does not show all quota information
When displaying quota information, the Management Console will not show extremely large
values in the “percent used” column. To see these values, use the mx quota showuser
command. You can also increase the quota for the affected user or group to reduce the
“percent used” value.
10902
HP StorageWorks Scalable NAS File Serving Software release notes 9