- Fixed bug: ipa incorrectly worked when some IPFW/IP6FW/IPF/PF
rule overflowed and this rule is not the first by order in
the corresponding parameter (in "ipa -t" output), it calculated
more bytes than actually should be calculated
- Fixed two bugs: ipa sometime incorrectly did accounting for
limits if statistics was subtracted in some rule
- Fixed bug: limit's start_time could be yyyy.mm.dd/24:00:00 in
the database, mktime(3) on tested systems understands such
local time and transforms it to next_day/00:00:00, now
start_time can't be 24:00:00 any more and always is
next_day/00:00:00.
- Fixed bug: if new_local_time - old_local_time > one_day (for
example as the result of date of ntpdate commands usage),
then ipa thought that a new day came, now it tries to find out
if local time is changed too quickly
PR: 59209
Submitted by: Andrey Simonenko <simon@comsys.ntu-kpi.kiev.ua> (maintainer)
b8ef07cc22