Problem Description:
A mysqld_safe instance is started. An InnoDB crash recovery begins which takes
few seconds to complete. During this crash recovery process happening, another
mysqld_safe instance is started with the same server startup parameters. Since
the mysqld's pid file is absent during the crash recovery process the second
instance assumes there is no other process and tries to acquire a lock on the
ibdata files in the datadir. But this step fails and the 2nd instance keeps
retrying 100 times each with a delay of 1 second. Now after the 100 attempts,
the server goes down, but while going down it hits the mysqld_safe script's
cleanup section and without any check it blindly deletes the socket and pid
files. Since no lock is placed on the socket file, it gets deleted.
Solution:
We create a mysqld_safe.pid file in the datadir, which protects the presence
server instance resources by storing the mysqld_safe's process id in it. We
place a check if the mysqld_safe.pid file is existing in the datadir. If yes
then we check if the pid it contains is an active pid or not. If yes again,
then the scripts logs an error saying "A mysqld_safe instance is already
running". Otherwise it will log the present mysqld_safe's pid into the
mysqld_safe.pid file.
Problem Description:
A mysqld_safe instance is started. An InnoDB crash recovery begins which takes
few seconds to complete. During this crash recovery process happening, another
mysqld_safe instance is started with the same server startup parameters. Since
the mysqld's pid file is absent during the crash recovery process the second
instance assumes there is no other process and tries to acquire a lock on the
ibdata files in the datadir. But this step fails and the 2nd instance keeps
retrying 100 times each with a delay of 1 second. Now after the 100 attempts,
the server goes down, but while going down it hits the mysqld_safe script's
cleanup section and without any check it blindly deletes the socket and pid
files. Since no lock is placed on the socket file, it gets deleted.
Solution:
We create a mysqld_safe.pid file in the datadir, which protects the presence
server instance resources by storing the mysqld_safe's process id in it. We
place a check if the mysqld_safe.pid file is existing in the datadir. If yes
then we check if the pid it contains is an active pid or not. If yes again,
then the scripts logs an error saying "A mysqld_safe instance is already
running". Otherwise it will log the present mysqld_safe's pid into the
mysqld_safe.pid file.
Some shell interpreters do not support '-e' test
primary to construct conditions.
man test 1 (on S10)
...skip...
-e file True if file exists. (Not available in sh.)
...skip...
Hence, check for the existence of a file using
'-e' might result in a syntax error on such
shell programs.
Fixed by replacing it by '-f'.
mysqld_safe script did not heed MySQL specific environment variable
$UMASK, leading to divergent behavior between mysqld and mysqld_safe.
Patch adds an approximation of mysqld's behavior to mysqld_safe,
within the bounds dictated by attempt to have mysqld_safe run on
even the most basic of shells (proper '70s sh, not just bash
with a fancy symlink).
Patch also adds approximation of said behavior to mysqld_multi
(in perl).
(backport)
manual merge
mysqld_safe script did not heed MySQL specific environment variable
$UMASK, leading to divergent behavior between mysqld and mysqld_safe.
Patch adds an approximation of mysqld's behavior to mysqld_safe,
within the bounds dictated by attempt to have mysqld_safe run on
even the most basic of shells (proper '70s sh, not just bash
with a fancy symlink).
Patch also adds approximation of said behavior to mysqld_multi
(in perl).
TO DEFAULT IGNOR
The test in mysqld_safe for the presence of the --plugin-dir and assigning a default
value to it were performed before the actual argument parsing.
This is wrong, as PLUGIN_DIR mysqld_safe code also uses MY_BASEDIR_VERSION to
look for version specific plugin directory if present.
Fixed by moving the PLUGIN_DIR logic after the parse_arguments() call.
Added a trottling mechanism to mysqld_safe.
On every 5 sub-second restarts it will wait for a second before attempting again.
Note that this requires 'sleep' and 'date' command line utilities.
The code would adjust and will do nothing if any of them is not present.
starting via mysqld_safe
Rather than hardcode the plugin directory, enhance mysql_config
to fix plugin path when running a relocated install, and use it
to provide the plugin directory to mysqld_safe.
- Update/fix file layouts for each package type, add new types for
native package formats including deb, rpm and svr4.
- Build all plugins, including debug versions
- Update compiler flags to match current release
- Add missing @VAR@ expansions
- Install correct mysqclient library symlinks
- Fix icc/ia64 builds
- Fix install of libmysqld-debug
- Don't include mysql_embedded
- Remove unpackaged manual pages to avoid missing files warnings
- Don't install mtr's test suite
via mysqld_safe
Plugin dir was set to a hard-coded path instead of relative the base dir.
This patch fixes this by using a path relative the basedir instead of the
plugin directory indicated by the configuration.
Several problems addressed:
1. The maximum value for --open_files_limit on non-windows boxes
is now raised to UINT_MAX (the maximum possible without significant
changes in the code). The maximum value on windows is kept to be
2048 due to a known limitation (bug 24509).
2. mysqld_safe now supports --open_files_limit=xx in addition to
--open-files-limit=xx
3. mysqld_safe always passes through --open[_-]files[_-]limit
to the underlying mysqld. It used to pass it through only if it
the user running the script has access to the root directory or
there was an --user argument specified.
4. Fixed a prototype in my_file.c to match its counterpart in
the other #ifdef branch.
When building the script directory using a CMake-based build, both the
variables in config.h.cmake (including PLUGINDIR) and the variables in
CMakeList.txt (which includes pkgplugindir).
However, for autotools-based builds, only pkgplugindir is substituted,
which means that the plugin-path is not substituted.
This patch solves the problem by using pkgplugindir, which works on both
CMake-based and autotools-based builds, instead of PLUGINDIR.
installed on debug version of server
When starting a server using mysqld_safe with the option
--mysqld-version=debug, the debug version of the server
is started. When attempting to load a plugin that is not
debug-built, the load will fail because a non-debug built
plugin use, e.g., fastmutex while the debug-built server
uses safemutex.
To solve this problem, mysqld_safe will set the plugin-dir
to use a directory where debug-built plugins are placed.
This patch sets the plugin dir to <plugin-dir>/whatever,
when mysqld-version=whatever, and just <plugin-dir> otherwise.
using cmake option INSTALL_LAYOUT=STANDALONE would produce the layout as in
tar.gz or zip packages.
INSTALL_LAYOUT=UNIX will produce unixish install layout (with mysqld being in sbin subdirectory , libs in lib/mysql etc). This layout is used for RPM packages.
Subtle differences in both packages unfortunately lead to the need to recompile MySQL to use with other package type - as otherwise for example default plugins or data directories would be wrong set.
There are numerous other variables that allow fine-tuning packaging layout. (INSTALL_BINDIR, INSTALL_LIBDIR , INSTALL_PLUGINDIR etc).
This options are different from autotools as they do not expect full paths to directories, but only subdirectory of CMAKE_INSTALL_PREFIX.
There are 2 special options that expect full directory paths
- MYSQL_DATADIR that defines default MYSQL data directory (autotools equivalent
is --localstatedir)
- SYSCONFDIR can be added to search my.cnf search path (autotools equivalent is --sysconfdir)
(revision-id:guilhem@mysql.com-20090505113602-l12kxupeatve18dh).
Such bug led "mysqld_safe --core_file_size=#" to not work because mysqld_safe wouldn't recognize
that "ulimit -c" is needed; only --core-file-size=# worked. Same for --open_files_limit and other
options with _ where mysqld_safe needs to do something more than passing to mysqld.
Original fix by Erik Ljungstrom erik at ibiblio dot org ; slightly modified here.
Tested on all internally accessible Unix.
doesn't find 'logger'
Due to a variable quoting mistake, the $PATH environment
variable isn't parsed correctly when searching for the
existence of the desired executable(s) (logger in this
case).
This patch removes the quotes.
The fix for bug 28544 moved our package data from ./share/mysql
to ./share. mysqld_safe had the old directory hard-coded. The
fix is to use the @pkgdatadir@ and @prefix@ values, to adapt to
different ways of building the package.
Our web server has been restructured several times, and references
to it in our source code has stayed the same. This patch from Paul
DuBois updates all URLs to modern semantics.
Don't use syslog by default; user will have to request it explicitly with the --syslog option.
Use "sed -u" to get unbuffered output from sed, if it's supported.
Otherwise, don't use sed at all - don't strip the timestamp from mysqld messages.
Also, add new --syslog-tag=FOO option, which adds "-FOO" to the tag used when logging messages to syslog (i.e., mysqld-FOO or mysqld_safe-FOO)
Also, explicitly mention where log messages are going, so user can more easily find them.
Also, check if 'logger' is in the PATH, and log to the error log file if it can't be found.
Try find in order 'mysqld-nt', 'mysqld', 'mysqld-debug' and 'mysqld-max'
mysqld_safe.sh:
Don't try 'mysqld-max' before 'mysqld' if no server given, bug#17861
