update (OK agc@)
The EventLog library aims to be a replacement of the simple syslog() API
provided on UNIX systems. The major difference between EventLog and syslog
is that EventLog tries to add structure to messages.
Where you had a simple non-structrured string in syslog() you have a
combination of description and tag/value pairs.
EventLog provides an interface to build, format and output an event record.
The exact format and output method can be customized by the administrator
via a configuration file.
pulled in from ../../devel/glib2/modules.mk, but it's now needed
in sysutils/gio-fam/Makefile.
Not bumping PKGREVISION because the package did not build before.
one that's unnecessary.
version 0.7.24
=====================================
2011-03-17
* Add cdinfo_get_track_last_lsn() which is like cdinfo_get_track_last_lsn()
* Make libvcdinfo work when called from C++, add C++ example program
* Fail less severely when trying to determine if medium is VCD
* Documentation and doxygen improvements
* Guard against various srncat buffer overflows
* Correct copyright and GPL version indication a containingReed-Solomon encoding
* Patch for memory exhaustion for VCD 2.0 disks SR 105012
Based on PR#45262 by Brian De Wolf, some fixes by me.
The purpose of this module is to implement a buffered version of the select
interface that operates on lines, rather than characters. Given a set of
filehandles, it will block until a full line is available on one or more of
them.
Kernel modules allow the system administrator to dynamically add and
remove functionality from a running system. This ability also helps
software developers to develop new parts of the kernel without constantly
rebooting to test their changes.
This package provides the necessary utilities for using kernel modules on
NetBSD 5.x.
The modload utility loads a kernel module specified by the module parameter
into the running system. The modunload utility unloads a loadable kernel
module from a running system. The modstat utility displays the status of
any kernel modules present in the kernel.
Changes since version 1.0.29:
A bug fix in the handling of readdir errors; in earlier versions,
it was theoretically possible for a failing hard drive or other
errors in reading directories to result in files being silently
omitted from an archive.
Several bug fixes relating to the handling of @archive directives
with mtree files.
A bug fix to prevent cache directory corruption resulting in
tarsnap failing if it was interrupted at exactly the right
(wrong) moment in its operation.
A bug fix to correctly handle ~ in tarsnap -s path substitutions.
Many more minor bug fixes.
1.6.0
===
9404a7a (#7670) Add an acceptance test
0c23845 maint: Fix spelling of acceptance directory
926e912 (#7670) Stop preloading all facts in the application
2255abe (#7670) Never fail to find a fact that is present
8002c24 (#7507) Fix 1.9.2 test failure
0635822 Removed inappropriately uncredited Ohai method from ec2 fact
6b1cd16 (#6614) Update ipaddress6 fact to work with Ruby 1.9
21fe217 (#6612) Changed uptime spec to be endian agnostic
19f96b5 (#6728) Facter improperly detects openvzve on CloudLinux systems
5b10173 (#5135) Fix faulty logic in physicalprocessorcount
53cd946 Ensures that ARP facts are returned only on EC2 hosts
bfa038d Fixed#6974 - Moved to Apache 2.0 license
d56bca8 refactor the mechanism for allowing for resolution ordering to be influenced
9f4c5c6 (#6740) facter doesn't always respect facts in environment variables
7441b32 Partial fix for #6971 - Fix for virtual tests
7f3e89d (#2714) Fixed faulty test
bfc16f6 (#2714) Added timeout to prtdiag resulution
c2ff833 (#5135) Refactored physicalprocessorcount
0c4a98b Re-factor. Do not use pure-Ruby file reading against "/proc/cpuinfo" and possibly any entry under "/sys" from the sysfs file system.
cb52b06 Fix. Using sysfs file system entries to count the number of physical CPUs. Fall-back to "/proc/cpuinfo" included for backward-compatibility with legacy systems.
3efa9d7 (#3856) Add virtualbox detection via lspci (graphics card), dmidecode, and prtdiag for Solaris and corresponding tests. Darwin case is not handled yet.
7c80172 (#6883) Update Facter install.rb to be slightly more informative.
d31e3f9 (#5394) Document each Facter fact.
af4947c (#6862) Add a default subject for the mail_patches rake task
d6967a0 (#6613) Switch solaris macaddress fact to netstat
e056218 (#6817) Fix for Ruby 1.9 by calling .each_line on a string
861c2b2 maint: cleanup whitespace
f6c9927 (#6719) Corrected faulty logic in bugfix
e42e57c (#3856) Add virtualbox detection via lspci (graphics card), dmidecode, and prtdiag for Solaris and corresponding tests. Darwin case is not handled yet.
0b5b546 (#6883) Update Facter install.rb to be slightly more informative.
7c08270 (#5394) Document each Facter fact.
06eb3f5 (#6883) Update Facter install.rb to be slightly more informative.
1063753 (#6862) Add a default subject for the mail_patches rake task
56b5f10 (#6613) Switch solaris macaddress fact to netstat
fd4f31c (#6817) Fix for Ruby 1.9 by calling .each_line on a string
72996ff maint: cleanup whitespace
Upstream changelog:
## 2.8.0 / August 3 2011
A short release, after the last. Announcing Rails 3.1 asset pipeline support.
The asset pipeline support requires an additiona `load` in your `Capfile`.
You can see information pertaining to the pull request, including the inline
comments here: https://github.com/capistrano/capistrano/pull/35
Documentation will be available soon in the wiki.
* Drop-In Rails 3.1 asset pipeline support. (Chris Griego)
## 2.7.0 / August 3 2011
A fairly substantial release. There are fixes so that current_release works
during dry-runs, (although, apparently still not with bundler.)
The test-suite was also modified to work with Ruby 1.9.2, except in one case
where Ruby 1.9.x calls `to_ary` and `to_a` on mocks, which still makes an
error. 1.9.x has always been supported, but due to lack of maintenance on my
part the tests didn't ever pass.
The `start`, `stop` and `restart` tasks have been reduced to mere hooks into
which extensions can define their own functionality.
The `readme` was also slightly improved, simply tweaks to express how best to
run the test suite.
* Ensure dry-run works with `:current_release` variable (Carol Nichols)
* Added a new variable `:git_submodules_recursive`, setting the value to false
will ensure Git doesn't recursively initialize and checkout submodules. (Konstantin Kudryashov)
* Added an additional task option, `:on_no_matching_servers`, setting the
value to `:continue` will ensure tasks with no matched servers continue
without error, instead of raising `Capistrano::NoMatchingServersError` as was
the previous behaviour. (Chris Griego)
A huge thanks to all contributors, as always!
Remember: @capistranorb on twitter for news.
## 2.6.1 / June 25 2011
A short maintenance release, Some fixes to the verbose flag inside the Git SCM
as well as another argument for the (internal) `variable()` command, offering
a default. The Git SCM is now verbose by default, but can be disabled by
setting `:scm_verbose` to false.
There has been an additional method added to string, within the context of the
test suite, I'm always sketchy about adding additional methods to core
classes, but it's a short term fix until I make the time to patch the test
suite not to compare strings literally. The method is `String#compact`, and is
implemented simply as `self.gsub(/\s+/, ' ')`.
Here's the run-down of changes, and their committers, as always - a huge thank
you to the community that continues to drive Capistrano's development.
* `deploy:setup` now respects `:group_writable` (Daniel Duvall)
* Fixes to `:scm_verbose` for the Git module (defaults to On.) (Matthew Davies)
* Will now copy hidden files in the project's root into the release
directory (Mark Jaquith)
* Now handles closing already-dead connections in a sane way (does not raise
an exception) (Will Bryant)
* Renamed `Capistrano::VERSION::TINY` to `Capistrano::VERSION::PATCH` (Lee
Hambley)
* Removed the `VERSION` file (Lee Hambley)
Upstream changelog:
=========
* :release:`1.2.0 <2011-07-12>`
* :feature:`22` Enhanced `@task <fabric.decorators.task>` to add :ref:`aliasing
<task-aliases>`, :ref:`per-module default tasks <default-tasks>`, and
:ref:`control over the wrapping task class <task-decorator-and-classes>`.
Thanks to Travis Swicegood for the initial work and collaboration.
* 🐛`380` Improved unicode support when testing objects for being
string-like. Thanks to Jiri Barton for catch & patch.
* :support:`382` Experimental overhaul of changelog formatting & process to
make supporting multiple lines of development less of a hassle.
* :release:`1.1.2 <2011-07-07>` (see below for details)
* :release:`1.0.2 <2011-06-24>` (see below for details)
Prehistory
==========
The content below this section comes from older versions of Fabric which wrote
out changelogs to individual, undated files. They have been concatenated and
preserved here for historical reasons, and may not be in strict chronological
order.
----
Changes in version 1.1.2 (2011-07-07)
=====================================
Bugfixes
--------
* :issue:`375`: The logic used to separate tasks from modules when running
``fab --list`` incorrectly considered task classes implementing the mapping
interface to be modules, not individual tasks. This has been corrected.
Thanks to Vladimir Mihailenco for the catch.
Dragonfly users, please test that is still builds !
pkgsrc change: as snmp support doesn't depend on net-snmp
anymore, remove snmp option and always build snmp support.
cgi option now depend on graphics/gd/
User-visible changes sinces 3.14.3:
2010-08-30 14:18 adk0212
* src/drivers/snmplite/: mge-mib.cpp, mge-oids.h, mibs.cpp:
Add support for MGE SNMP MIB. Contributed by Lars Täer
<taeuber@bbaw.de>
2010-07-30 18:04 adk0212
* src/drivers/snmplite/: apc-mib.cpp, apc-oids.h, mib.cpp, oids.h,
rfc1628-mib.cpp, rfc1628-oids.h, snmplite.cpp, snmplite.h:
Add support for RFC1628 SNMP MIB. Refactor APC MIB and create
MibStrategy struct for associating MIB/CI mapping with
corresponding processing function. RFC1628 strategy is coded per
the MIB but untested.
2010-01-10 10:29 adk0212
* include/defines.h, src/apctest.c, src/drivers/usb/usb.c:
Add apctest support for reading/setting self-test interval on USB.
Also show current setting in UPS status. Contributed by James
Belleau <jpbelleau@gmail.com>
2009-10-25 11:03 adk0212
* configure, autoconf/config.h.in, autoconf/configure.in,
autoconf/variables.mak.in, include/struct.h,
platforms/etc/apcupsd.conf.in, src/drivers/Makefile,
src/drivers/drivers.c, src/drivers/snmplite/Makefile,
src/drivers/snmplite/asn.cpp, src/drivers/snmplite/asn.h,
src/drivers/snmplite/mib.cpp, src/drivers/snmplite/snmp.cpp,
src/drivers/snmplite/snmp.h, src/drivers/snmplite/snmplite.cpp,
src/drivers/snmplite/snmplite.h, src/lib/apcconfig.c,
src/lib/apcstatus.c:
Add SNMP Lite driver which does not depend on net-snmp library.
This makes it more portable and eliminates need to move libsnmp.so
to /lib in order to do a killpower on systems where /usr is
unmounted.
2009-09-01 20:30 adk0212
* src/apctest.c:
Implement battery calibration in apctest for USB models. Thanks to
James Belleau <james@belleau.net> for the original implementation
which has been modified somewhat in this commit.
2009-05-02 10:30 adk0212
* src/action.c:
Change log level of UPS self-test messages to WARNING from ALERT.
Given that self-test messages are routine, they do not belong at
LOG_ALERT. Contributed by Dave Ewart <davee@ceu.ox.ac.uk>.
2009-04-25 10:58 adk0212
* src/lib/apcconfig.c:
Remove EVENTFILE, EVENTFILEMAX config directives. These were
replaced by the plural versions that are in use today (EVENTSFILE,
EVENTSFILEMAX) almost 10 years ago. It's time to kill the old
names. (h/t Trevor Roydhouse <trev@sentry.org>)
2009-03-02 17:48 adk0212
* doc/apcupsd.man, include/drivers.h, include/extern.h,
src/action.c, src/apctest.c, src/apcupsd.c, src/device.c,
src/options.c, src/drivers/drivers.c,
src/drivers/apcsmart/apcsmart.h, src/drivers/apcsmart/smart.c,
src/drivers/apcsmart/smartoper.c, src/drivers/usb/usb.c,
src/drivers/usb/usb.h, src/lib/apclock.c:
Add support for turning the UPS off completely. This complements
existing hibernate (aka killpower) functionality. Turn-off is
implemented for apcsmart and USB drivers, subject to support for
the relevant commands in the UPS itself. Contributed by Keith
Campbell <campbell@econnectix.com>.
2008-06-29 11:12 adk0212
* src/action.c:
Fix bug in LOWBATT glitch handling. We must examine LOWBATT for
changes every time thru the status loop, not just during the
OnBattery state. Otherwise we can miss the initial LOWBATT
assertion, which defeats the glitch rejection logic.
2008-05-06 20:16 skoona
* src/gapcmon/gapcmon.c:
Corrected the use of NOMPOWER and the calc of current usage amount
2008-05-04 11:13 adk0212
* src/drivers/usb/usb.c:
Add a heuristic to fix up incorrect NOMINV or NOMOUTV. Some UPSes
(RS 500) report decivolts instead of volts. Reported by Kirill S.
Bychkov <yason@linklevel.net>.
2008-01-27 12:00 adk0212
* include/struct.h, platforms/etc/apcupsd.conf.in,
platforms/mingw/apcupsd.conf.in, src/device.c,
src/drivers/net/net.c, src/drivers/snmp/drv_powernet.c,
src/lib/apcconfig.c:
Add POLLTIME directive to control UPS polling interval. NETTIME is
accepted as a synonym for compatibility with old config files.
