HP StorageWorks Clustered File System 3.6.0 Windows Storage Server Edition (software-only) Release Notes (432086-004, February 2008)
F ilesy stem resize does not work correctly with mount points [7591]
The HP Management Console includes an option to extend the size of a fi lesystem. After a filesystem is
extended, HP Clustered File System attempts to reassign the drive letters or mount p oints associated with
the
filesystem. HP Clustered File System reassigns drive letters correctly; however, it fails to reassign mount
points. Re-create the appropriate mount points manually after the extend operation is complete.
To re-create th
emount points, first manually delete the old mount points (the empty directories on the local
NTFS volumes of each node) . Next, recreate the empty directories and then use the HP Management
Console to reassign the mount points m anually.
Virtual host can cause route change [8290]
When a virtual host is instantiated, it uses the network m ask of the interface upon which it is activated.
This can cause an unexpected route change if the virtual host address was intended for some other
interface on another network, or if the virtual host address is mistyped.
For example, the following command adds the 10.10.1.100 address on the 10.11.1.1 interface:
mx vhost add 10.10.1.100 10.11.1.1
If the 10.11.1.1 interface had a netmask of 255.255.0 .0, then the traffic for the 10.10/16 subnet would
be
directly routed to the 10.11/16 network for this server, rather than being routed to a gateway on
the
10.11/16 network.
URL in HTTP service monitor can cause virtual host to fail [8335]
The URL assigned to an HTTP service monitor should not reference the virtual host with which the
HTTP service monitor is associated. If the virtual host is down, the HTT P service monitor cannot make
a
connection to the virtual host address and will also be down. A down service monitor on a virtual
host
then t
ends
to keep the virtual host from ever being instantiated.
psfscheck incorrectly reports attribute errors [9 3 2 6]
When psfsch eck is run on a filesystem created on HP Clustered File SystemGateway 2.5 or earlier, it may
report an error s uch as the following:
Pass 3: Checking semantic tree
directory 0 8334 has broken attribute 0x0
You c an safely ignore this message. The attribute bit was not set in HP Clustered File SystemGateway 2.5
and
earlier releases. (If psfscheck is run in a m o de other than read-only, it fixes the error.)
Issues w
hen
installing from a CIFS share [9442, 9480]
HP Clust
ered File System can be installed from a CIFS share; however, note the following issues:
• HP Clustered File System can be installed via Terminal Services from a CIFS share only on
Windows 2003. If you are installing on Windows 2000 using Terminal Services, install either
from CD or from a local copy of the software.
• When th
e
mxcheck utility is executed from a CIFS share, the HBA test fails. Although a detailed
error m
essage is not reported, a line such as the following prints:
Failure: Check for supported HBAs, drivers, and settings (hba_test)
TheHBA testpasseswhen mxcheck is executed from CD or from a copy on a local d isk.
Cluster may not start when fencing method is changed [12050]
If the fencing method is changed without shut ting down all the servers in the cluster cleanly, the cluster
may not be able to start successfully. To check for and correct this condition, run the mpdump command
with no o p tions.
HP StorageWorks Clustered File System 3.6.0 WSS Edition (soft ware-only) release notes
9