HP Tru64 UNIX and TruCluster Server Version 5.1B-5 Patch Summary and Release Notes (March 2009)
• Corrects a system panic when running with big pages enabled.
• Addresses problems that occur when taking noninteractive core dumps using the
coredump command.
• Corrects potential hangs of applications using the pshared subsystem that can
occur as a result of a thread failing to wake up after its condition variable timer
had expired.
• Fixes two potential problems in the NFS V3 client in which unstable writes can
remain uncommitted when they should have been committed to stable storage.
• Removes erroneous "No B-cache detected" messages from certain configurations.
• Prevents a potential panic that generates the message “memory_test=partial” or
“memory_test=none.”
• Allows the stat system call to correctly report the st_blocks on a CD-ROM file.
• Improves performance for applications with large in-core data sets
• Corrects a system panic caused by stack growth.
• Adds the ability to join more than 20 IP multicast groups on a given socket.
• Fixes a multiple process hang (which cannot be terminated with Ctrl/c) that can
occur if process A attempts to attach process B to another RAD while at the same
time, process B attempts to attach process A to another RAD (a classic deadlock).
• Changes the implementation of the NFS server's duplicate request cache from a
statically allocated monolithic entity to a dynamically allocated entity.
• Corrects a kernel memory fault panic under certain heavy system loads when
using the /proc file system to debug processes.
• Corrects a problem in which using PRSABORT in the /proc file system does not
correctly abort a system call in the process being debugged.
• Fixes a "simple_lock: time limit exceeded" panic involving the vm_object.ob_lock
lock.
• Fixes a rare case of a thread blocking when waiting for memory.
• Fixes a problem in which some IP fragments of NFS over UDP read replies may
be sent to the wrong MAC address.
• Permits the setting of the setuid, setgid and sticky bits on NFS served files.
• Fixes kernel memory faults that can occur when packetfilter support is used
incorrectly as a result of the following:
— A misconfiguration of the packetfilters system configuration parameter in the
net subsystem.
— The dynamic load of the packetfilter subsystem on a system that has packetfilter
statically built into the kernel.
• Fixes a condition in which the fork() function returns EAGAIN incorrectly on a
process using gh_chunks.
• Corrects a problem in which interface Unicast packet counters seem to go
backwards when retrieved via SNMP.
3.4 Summary of Base Operating System Patches 131