Hello! sys-power/apcupsd-3.14.2 has been released over one month ago ... The apcupsd acpcontrol patch need to be modified, as its not anymore working.
The apccontrol patch is no more needed.
Apcupsd 3.14.2 is the latest stable release, containing many bug fixes and new features over the previous 3.12.x stable series. Users of all previous versions are encouraged to upgrade. Release notes: http://sourceforge.net/project/shownotes.php?group_id=54413&release_id=541283 BUG FIXES * USB fix for BackUPS LCD series. This fix carries a risk of breaking other models, but test results are positive so far. * Fix linux-usb device detection bug when multiple hiddev nodes are present. Fixes Fedora Bugzilla #245864: <https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=245864> * Fix SNMP port selection when using SNMP driver * Add workaround for broken BackUPS Pro model that returns certain reports encoded in ASCII. Based on a patch from David Fries <David@Fries.net>. * apcsmart: Wait 20 seconds before declaring COMMLOST. This is necessary when using a smart serial cable on an UPS with an SNMP monitoring card installed. The SNMP card appears to disrupt serial comms for several seconds when interesting events are in progress. * Documentation updates: Fix up udev info for modern Linux distros. Update Windows section for apctray. * Don't leak unnecessary descriptors to child when execing apccontrol. * Fix bug with multi-threaded net_open() calls. net_open() was using a global buffer for the destination address. Yikes! Bug manifestied itself in apctray with multiple monitors occasionally (and timing-dependently) pointing to the wrong UPS. * Correct multimon.conf humidity monitoring. Problem fixed by Patrick Frei. Reference: https://bugzilla.novell.com/show_bug.cgi?id=199834 * Fix race between attach_ups() and detach_ups() which could result in the UPSINFO structure being freed prematurely. Thanks to James Abbott <j.abbott@imperial.ac.uk> for running many tests to help me find this bug. * Remove dependency on libc++ (again). Must use gcc as linker when using libsupc++ since g++ automatically links against full libc++. Guess we should add this to portage soon.
Related to #65211
In cvs.
