STABILITY FIX:
Fixed double free'd pointer bug in mysql_change_user() handling, that
enabled a specially hacked version of MySQL client to crash mysqld.
Note, that one needs to login to the server by using a valid user
account to be able to exploit this bug.
to powerpc and also affects i386/sparc. the exact conditions that
cause this to happen are unknown at the moment though this will
have to be investigated in closer detail.
this port on sparc.
- Set CXX=${CC} to get rid of libstdc++ in order to improve stability
under heavy load; From: FreeBSD.
- Build mysqld statically by default, enable InnoDB by default too.
- Replace max FLAVOR with bdb FLAVOR and only enable it on i386 for now
until it can be tested on other archs.
- Add comment to MySQL to indicate that this install of MySQL is from the
OpenBSD port/package.
- Add RUN_DEPENDS on p5-Msql-Mysql for some of the Perl-based MySQL
scripts.
- Set the mysql account to be apart of the daemon login class.
--
- The port has switched back to using native threads. Only tested and known to work on i386. For everything else the port is currently marked BROKEN until the appropriate architectures are tested. Do not even try this if you're not running -current libc_r.
- As suggested by Marc Espie, make symlink for the MySQL client library in lib/ to eliminate the need for adding lib/mysql to the shlib dir path. Porters have to be careful and ensure autoconf scripts do not arbitrarily enable MySQL support after this change.
individual port, introduce a new variable for bsd.port.mk, VMEM_WARNING.
By setting this variable to `Yes', the user is given a warning about the
virtual memory requirements just before the pre-build stage.
o Also give some examples for different shells for what to do about
it, and mention login.conf(5). bash, for example, has a confusing
-v flag to ulimit.
o Document this in bsd.port.mk.5
espie@ ok
