openafs (1.6.6~pre2-1) UNRELEASED; urgency=low
* New upstream pre-release.
+ - Linux kernels up to 3.12 are now supported, including kernels with
+ user namespace support enabled (which affects Debian's 3.12-1 kernel
+ and newer).
+ - Fixed core dumps into AFS with current kernels.
+ - When starting the client fails, backing device information created
+ in sysfs is now properly cleared.
+ - The AFS mountpoint specified in the cacheinfo file must now be an
+ absolute path.
+ - Stop tracking file locks on read-only volumes. Write locks always
+ fail and read locks always succeed.
+ - New fs flushall command to discard all cached data.
+ - Fixed a bug that could cause the client to incorrectly believe its
+ cache was up to date.
+ - New -rxmaxfrags switch to afsd to limit the number of UDP fragments
+ sent or received per RX packet.
+ - Fixed afsd threads entering an infinite loop.
+ - The file server now ignores any vice partitions with a NeverAttach
+ flag file present in the root directory.
+ - Enabled server-side NAT pings to refresh NAT timeouts.
+ - Forcing file server CPS recalculation (for IP ACLs) is now
+ restricted to administrators.
+ - vos examine of a volume in a transaction is now shown as busy again
+ rather than off-line.
+ - Multiple bug fixes to the salvager.
+ - Fixed a bug that could cause state information to be discarded when
+ restarting a large or busy file server.
+ - Fixed a vlserver bug during file server address registration.
+ - volserver supports a new -preserve-vol-stats option, which preserves
+ access statistics across volume restore and reclone operations.
+ - Releasing a volume after adding a new RO site no longer touches the
+ existing RO sites if the volume has not changed since the last
+ release.
+ - Fixed undefined ptserver behavior with too many allocated PTS ids.
+ - Avoid redefining assert in public header files.
+ - Documentation, diagnostics, and error message improvements.
-- Russ Allbery <rra@debian.org> Sat, 28 Dec 2013 11:00:17 -0800