HP StorageWorks Clustered File System 3.6.0 Windows Storage Server Edition (software-only) Release Notes (432086-004, February 2008)

Windows applica
tions
You may notice the following differences from NT FS filesystems when running applications on PSFS
filesystems:
• Certain Window
s
applications disable NTFS-related features when run on a PSFS filesystem.
For example, old er versions of Robocopy do not copy security descriptors. (As a workaround,
use Xcopy to copy the security descriptors.)
Open issues and workarounds for HP Clustered File System
The following open issues affect HP Clustered File SystemGateway operations.
Network inte
rfaces are marked “Up” and then “Down” repeatedly [3636]
The status of network interfaces may change between “Up” and “Down” about every five seconds when
there is only
one
server in the cluster and no other IP address is available to ping. This situation can
occur when a
two-node cluster loses a node, or the node is removed from service. However, the p roblem
can
occur in laboratory or test conditions. To avoid the problem, add a router to the PAN network.
The problem appears only if the only gateway ad d ress to ping is the interface IP address (that is, there is
not a
route t
o
an outside network). This scenario is unlikely to occur in practice, as there will likely be
many clients connected to the clustered server using a client network.
However, the problem can occur in laboratory or test conditions. To avoid the problem, add a router to
the
PAN netw
ork.
Management Console does not display filesystems created when HP Clustered
File SystemGateway was not running [3793]
If the HP Clustered File SystemGateway service is not running while a PSFS fi lesystem is created with the
psfsformat -f command, the filesystem d oes not display on the HP Management Console.
Workaround—Start the H P Clustered File SystemGateway ser vice before creating the filesystem.
Filesystem mounts may disappear temporarily while a node is booting [4301]
While a node in the cluster is booting, the HP Management Console can show filesystem mounts
disappearing and then reappearing on other nodes. This is an artifact of the cluster responding to a
new node
joining the cluster and causes no harm.
SNMP dies on QLogic SANbox2 FC switches [5 519]
In a cluster with a QLogic SANbox2 FibreChannel switch, SNMP can die periodically. When this occurs,
servers cannot be fenced, and servers that have already been fenced cannot unfence themselves.
To recover, go to Admin mode on the switch and then run the hotreset command:
admin start
admin> hotreset
The hotreset command resets the switch. (Before the switch is reset, all running applications save their
FibreChannel state. That information is then used by the applications when the switch comes back up.)
If the cluster does not recover on its own af ter the switch is reset, you may need to reboot the affected
servers or restart HP Clustered File Sy s temGateway on them.
HP StorageWorks Clustered File System 3.6.0 WSS Edition (software-only) release notes
7