Building MySQL and MariaDB on Windows using Visual Studio 2012 Express versus Visual Studio 2010 Express

April 5th, 2013

I mentioned in a recent post (and another recent, related post) I was going to try Visual Studio 2012 Express (as opposed to VS 2010 Express) for building MySQL 5.6 and MariaDB 10.0. on Windows.

I followed the exact same steps, but MySQL 5.6.10 would not build without fatal errors in 2012 (whereas it worked fine with VS 2010). Unfortunately, the fail errors were not too descriptive, so not quite sure what to do yet to fix it. Here is a brief snippet for those interested:

...
Done Building Project "C:\...\bld\sql\udf_example.vcxproj" (default targets).
Done Building Project "C:\...\bld\ALL_BUILD.vcxproj" (default targets) -- FAILED.
Done Building Project "c:\...\bld\package.vcxproj" (default targets) -- FAILED.
...
Build FAILED.
...

"c:\mysql\mysql-5.6.10\bld\package.vcxproj" (default target) (1) ->
"C:\mysql\mysql-5.6.10\bld\ALL_BUILD.vcxproj" (default target) (2) ->
"C:\mysql\mysql-5.6.10\bld\plugin\audit_null\audit_null.vcxproj" (default target) (18) ->
"C:\mysql\mysql-5.6.10\bld\sql\mysqld.vcxproj" (default target) (19) ->
(Link target) ->
  mysqld.exp : error LNK2001: unresolved external symbol
  _xmm@7fffffffffffffff7fffffffffffffff [C:\mysql\mysql-5.6.10\bld\sql\mysqld.vcxproj]
  C:\mysql\mysql-5.6.10\bld\sql\RelWithDebInfo\mysqld.exe : fatal error LNK1120:
  1 unresolved externals [C:\mysql\mysql-5.6.10\bld\sql\mysqld.vcxproj]

    6 Warning(s)
    2 Error(s)

Time Elapsed 00:01:06.67

On the plus side, MariaDB 10.0.1 compiled just fine with both Visual Studio 2012 Express and Visual Studio 2010 Express! So I was really happy about that. :)

I thought I could just keep both VSs installed, but they don’t seem to want to play nice, even when specifying the following for the MySQL 5.6 build (perhaps that’s just me though):

cmake .. -G "Visual Studio 10"

So, if you are building your own on Windows, then you may want to keep the above in mind (my tests have been on Windows 7 64-bit, fwiw).

I’ll continue to test, and post anything useful if there’s something worth mentioning.

 

Related Build Links:

 

Tags: , , , , , , , , , , , , , , , , , , , , ,
Posted in MySQL, SkySQL | 2 Comments »

Awesome to see the MySQL Ecosystem Flourishing

March 30th, 2013

I just wanted to take a moment and thank, notice, what ever you want to call it, but share some love with all those and all things MySQL.

I read the post Let’s Celebrate MySQL 5.6 GA! – MySQL Community Reception by Oracle by Oracle MySQL Group and it got me to thinking of how proud I am of (and proud to be a part of) the whole MySQL Ecosystem.

We *should* all celebrate MySQL 5.6 GA! I well remember the 3.22 and 3.23 days, and there were many folks before me already using MySQL!!!

I love to see how it has continued to grow, the ecosystem and all things MySQL, that is. MySQL is better than ever. MariaDB is better than ever. Percona Server is better than ever. You have great Support options with MySQL/Oracle, SkySQL/MariaDB, and Percona as well – not to mention numerous others. I just love the fact that everyone is flourishing, and improving, and the technology continues to advance and most importantly, everyone is working together to provide a great software and service to all MySQL/MariaDB/InnoDB/XtraDB/etc-related users out there.

Not to mention the monitoring softwares, the various plugins and connectors, the numerous storage engines, the number of 3rd party apps & helpful scripts, User Conferences, local meet-ups, books, tutorials, the number of Community bloggers who freely and spiritedly share their knowledge, time, advice, ideas, concerns, and so on. It’s just a great Community to be a part of, and I’m thrilled to see how it’s grown, and continues to grow over the years. And that’s just a testament of the quality of all you great people out there involved and helping. :)

I won’t get too mushy, but it is great to see the MySQL Ecosystem and all of those in it flourish, and may it continue for many, many years to come! :)

Big thanks to everyone out there contributing! “It takes a village, and we are that village”, as my wife always says. :)

 

Tags: , , , , , ,
Posted in MySQL, SkySQL | Comments Off

How to Build MySQL 5.6 on Windows from Source Code

March 28th, 2013

I just set up a new Windows 7 machine, and built MySQL 5.6 for the first time, so I wanted to share my steps for anyone interested.

** Prerequisites **
1. Install Microsoft Visual Studio (I installed VS 2010 Express – free)
2. Install cmake
3. Install bison (make sure no spaces in path)

** Build Instructions **

cd c:\mysql\mysql-5.6.10
mkdir bld
cd bld
cmake ..
cmake --build . --config relwithdebinfo --target package

(You don’t technically need the ‘bld’ dir, but I prefer it to keep the main dir ‘clean’. Then if you need to re-build, you can quickly just delete the whole ‘bld’ directory, and start fresh.)

Here is my actual output for those interested:

c:\>cd c:\mysql\mysql-5.6.10
c:\mysql\mysql-5.6.10>mkdir bld
c:\mysql\mysql-5.6.10>cd bld
c:\mysql\mysql-5.6.10\bld>cmake ..
-- Building for: Visual Studio 10
-- Running cmake version 2.8.10.2
-- The C compiler identification is MSVC 16.0.30319.1
-- The CXX compiler identification is MSVC 16.0.30319.1
-- Check for working C compiler using: Visual Studio 10
-- Check for working C compiler using: Visual Studio 10 -- works
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Check for working CXX compiler using: Visual Studio 10
-- Check for working CXX compiler using: Visual Studio 10 -- works
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Check size of void *
-- Check size of void * - done
-- MySQL 5.6.10
-- Packaging as: mysql-5.6.10-win32
-- Looking for include file sasl/sasl.h
-- Looking for include file sasl/sasl.h - not found
-- Looking for include file sys/devpoll.h
-- Looking for include file sys/devpoll.h - not found
-- Looking for include file signal.h
-- Looking for include file signal.h - found
-- Looking for include file sys/devpoll.h
-- Looking for include file sys/devpoll.h - not found
-- Looking for include file sys/epoll.h
-- Looking for include file sys/epoll.h - not found
-- Looking for include file sys/event.h
-- Looking for include file sys/event.h - not found
-- Looking for include file sys/queue.h
-- Looking for include file sys/queue.h - not found
-- Looking for TAILQ_FOREACH
-- Looking for TAILQ_FOREACH - not found
-- Looking for include file pthread.h
-- Looking for include file pthread.h - not found
-- Found Threads: TRUE
-- Looking for _aligned_malloc
-- Looking for _aligned_malloc - found
-- Looking for _aligned_free
-- Looking for _aligned_free - found
-- Looking for fedisableexcept
-- Looking for fedisableexcept - not found
-- Looking for posix_memalign
-- Looking for posix_memalign - not found
-- Looking for ntohll
-- Looking for ntohll - not found
-- Looking for clock_gettime
-- Looking for clock_gettime - not found
-- Looking for epoll_create
-- Looking for epoll_create - not found
-- Looking for epoll_ctl
-- Looking for epoll_ctl - not found
-- Looking for inet_ntop
-- Looking for inet_ntop - not found
-- Looking for kqueue
-- Looking for kqueue - not found
-- Looking for kqueue
-- Looking for kqueue - not found
-- Looking for signal
-- Looking for signal - found
-- Looking for timeradd
-- Looking for timeradd - not found
-- Looking for timerclear
-- Looking for timerclear - not found
-- Looking for timercmp
-- Looking for timercmp - not found
-- Looking for timerisset
-- Looking for timerisset - not found
-- Check size of time_t
-- Check size of time_t - done
-- Performing Test TIME_T_UNSIGNED
-- Performing Test TIME_T_UNSIGNED - Failed
-- Looking for sched_getcpu
-- Looking for sched_getcpu - not found
-- Looking for asprintf
-- Looking for asprintf - not found
-- Using cmake version 2.8.10.2
-- Not building NDB
-- Googlemock was not found. gtest-based unit tests will be disabled. You can run cmake . -DENABLE_DOWNLOADS=1 to automatically download and build required components from source.
-- If you are inside a firewall, you may need to use an http proxy: export http_proxy=http://foo.bar.com:80
-- Cannot find wix 3, installer project will not be generated
-- Configuring done
-- Generating done
-- Build files have been written to: C:/mysql/mysql-5.6.10

c:\mysql\mysql-5.6.10\bld>cmake --build . --config relwithdebinfo --target package
...
... tons of build/compilation output ...
...
    406 Warning(s)
    0 Error(s)

Time Elapsed 00:16:46.65

There you go!

The zip package (mysql-5.6.10-win32.zip in this case) will be located in bld/.

I will say that is a lot of warnings, though I’m fairly used to them on Windows. However, I plan to try VS 2012 Express to see if that makes any difference.

Hope this helps. :)

 

Tags: , , , , , , , , ,
Posted in MySQL, SkySQL | Comments Off

How to Build MariaDB 10.0 on Windows from Source Code

March 26th, 2013

I just wanted to share my steps for building MariaDB 10.0 on Windows (10.0.1 in this case):

cd maria-10.0.1
mkdir bld
cd bld
cmake .. -G "Visual Studio 10"
cmake --build . --config relwithdebinfo --target package

That’s it! The zip file is created right there in bld/.

Hope this helps. :)

 

Tags: , , , , , , , , , , ,
Posted in MySQL, SkySQL | 1 Comment »

SkySQL & MariaDB’s 2nd Annual MySQL & Cloud Database Solutions Day

March 26th, 2013

As the press release says: “Come have a (free) beer with Monty: SkySQL & MariaDB’s 2nd Annual MySQL & Cloud Database Solutions Day”!

“Are you planning on being in Santa Clara for Percona Live: MySQL Conference and Expo in April? If so, don’t let the learning and networking stop on Thursday!

We invite you to join SkySQL & MariaDB for our 2nd Annual MySQL® & Cloud Database Solutions Day, Friday, April 26, 2013 at the Hyatt Regency Santa Clara. This event will be taking place directly after the Percona Live event.

During this free one-day event, you will receive hands-on training and consulting on MySQL and MariaDB database solutions from the experts at SkySQL, MariaDB, and their partners.”

Speakers will include Big Data, Analytics and Cloud experts, as well as SkySQL customers and MariaDB users. Including:

Monty Widenius – MariaDB
Yoshinori Matsunobu – Facebook
Seppo Jaakola – Codership
Kyle MacDonald – OpenStack Foundation
Full Speaker List

“The event will feature two technical tracks: one focused on the MySQL database and related technologies, and the second on best practices for deploying database applications in the Cloud.”

To learn more, take a look at our event schedule, featured speakers, and venue and please register today:

Register Today
Full Schedule

 

Tags: , , , , ,
Posted in MySQL, SkySQL | Comments Off

Enabling the Verbose InnoDB Lock Monitor in MariaDB and Percona Server for XtraDB+ and XtraDB

March 25th, 2013

I enabled the InnoDB Lock Monitor in my MariaDB 5.5 instance (using XtraDB+ as the InnoDB – which is the default in MariaDB) and noticed that while the SHOW ENGINE INNODB STATUS was being logged to the error log, it wasn’t logging the “additional” lock information – it just looked like the plain ‘ole INNODB STATUS.

Long story short, Percona added a new variable so one has better control over what gets logged:

innodb_show_verbose_locks

If off (default), then the InnoDB Lock Monitor logs the normal INNODB STATUS, and if enabled, then it logs it with the extended lock information.

They also created another variable that goes along with this one (and the InnoDB Lock Monitor), which is:

innodb_show_locks_held

This variable indicates the number of locks to print that are held for each InnoDB transaction (the default is 10, max is 1000).

For reference, these 2 options are discussed further in Percona’s manual.

Hope this helps. :)

 

Tags: , , , , , , , , , ,
Posted in MySQL, SkySQL | Comments Off

Dealing with the dreaded “The printer is out of paper” or ERROR_OUT_OF_PAPER Error on Windows ;)

March 20th, 2013

Caution: This is just a fun post. :D

I was performing some testing on Windows the other day and encountered the following error in the error log:

130313  9:40:03 [ERROR] Error writing file 'G:\mysql\log.txt' (errno: 28)

Quickly, I ran “perror 28″ from the command line:

c:\>perror 28
Win32 error code 28: The printer is out of paper.

Hehe, really, Windows?!?

In fact, the drive was out of space. No more could be written to the general log, hence the error.

Fwiw, the Linux OS code 28 reports what you would expect:

OS error code  28:  No space left on device

So, if you ever run into “The printer is out of paper” error, you’ll know what to do. ;)

 

Tags: , , , , , , , ,
Posted in MySQL, SkySQL | 4 Comments »

InnoDB Plugin Version Numbering in MySQL and MariaDB

March 18th, 2013

As some of you may or may not know, I’ve maintained a list of all InnoDB Plugin versions as they’ve historically contained a different version (entirely different numbering scheme) than the MySQL distribution they were included with.

This list was most helpful for troubleshooting various InnoDB issues when the plugin may (or may not) have been involved, and/or for benchmarking, etc. And it’s fair to say it was more useful when the InnoDB plugin was not the mainstream, which it is now.

However, with the latest releases, in MySQL and MariaDB, the “InnoDB Version” simply matches the “MySQL Version”. These “latest releases” include: MySQL 5.6.10, MySQL 5.5.30, MySQL 5.1.68, and MariaDB 5.5.30

Of course this isn’t the most newsworthy story, but having maintained this “list” the past couple/few years, I was happy to see the change, and at least wanted to mention it.

There won’t be a need for me to further update the page, but it is up-to-date now, so if you happen to need to know what version of InnoDB Plugin an older version of MySQL or MariaDB is using, then this page will still be there for you (it’s also in the right-hand column of my site under “Pages” titled “InnoDB Plugin Versions“).

Hope this helps. :)

 

Tags: , , , , , , , ,
Posted in MySQL, SkySQL | 1 Comment »

Troubleshooting MySQL Crashes related to Metadata Locking

March 15th, 2013

I just wrote an article about “Troubleshooting ‘Waiting for table metadata lock’ Errors for both MyISAM and InnoDB Tables” and then ran into a new, different metadata locking issue right after I posted it, and so I thought I’d share that too, just in case anyone ever encounters a similar situation.

In this case, mysqld kept crashing on restart, reporting “out of memory” errors:

/opt/app/mysql/product/mysql/bin//mysqld: Out of memory (Needed 840 bytes)
stack_bottom = 7fb4ebaeae58 thread_stack 0x40000
mysqld: /mysql/mysys/my_new.cc:52: int __cxa_pure_virtual():
  Assertion `! "Aborted: pure virtual method called."' failed.
mysqld: /mysql/mysys/my_new.cc:52: int __cxa_pure_virtual():
  Assertion `! "Aborted: pure virtual method called."' failed.
Fatal signal 6 while backtracing

There is a recent bug fixed (in 5.5.29) that mentions the same assert above, but that is not the same problem here, as that bug always reports “InnoDB: Failing assertion: page_get_n_recs(page) > 1″.

Looking further into these repeated crashes, I noticed the following some of the stack traces always showed this:

stack_bottom = 7f2cd227ce58 thread_stack 0x40000
/opt/app/mysql/product/mysql/bin//mysqld(my_print_stacktrace+0x35)[0x79e2d5]
/opt/app/mysql/product/mysql/bin//mysqld(handle_fatal_signal+0x403)[0x66e233]
/lib64/libpthread.so.0(+0xf7c0)[0x7f46041577c0]
/../mysqld(_ZN10MDL_ticket6createEP11MDL_context13enum_mdl_type+0x26)[0x61a3e6]
/../mysqld(_ZN11MDL_context21try_acquire_lock_implEP11MDL_requestPP10MDL_ticket+0xa2)[0x61c4c2]
/../mysqld(_ZN11MDL_context12acquire_lockEP11MDL_requestm+0x63)[0x61cb53]
/../mysqld[0x539d76]
/../mysqld(_Z10open_tableP3THDP10TABLE_LISTP11st_mem_rootP18Open_table_context+0x1b1)[0x541591]
/../mysqld(_Z11open_tablesP3THDPP10TABLE_LISTPjjP19Prelocking_strategy+0x42b)[0x542d2b]
/../mysqld(_Z30open_normal_and_derived_tablesP3THDP10TABLE_LISTj+0x48)[0x543678]
/../mysqld(_ZN18Prepared_statement7prepareEPKcj+0xc5e)[0x58a99e]
/../mysqld(_Z19mysqld_stmt_prepareP3THDPKcj+0xae)[0x58bbee]
/../mysqld(_Z16dispatch_command19enum_server_commandP3THDPcj+0xc21)[0x57c4e1]
/../mysqld(_Z24do_handle_one_connectionP3THD+0x167)[0x613447]
/../mysqld(handle_one_connection+0x54)[0x6134b4]
/../mysqld(pfs_spawn_thread+0x5b)[0x7f6abb]
/lib64/libpthread.so.0(+0x77b6)[0x7f460414f7b6]
/lib64/libc.so.6(clone+0x6d)[0x7f46033e7c5d]

At first glance, it’s kind of hard to read. However, if you focus in, you can extract some useful information.

Looking just before the crash (handle_fatal_signal), we see these 3 lines:

/../mysqld(_ZN10MDL_ticket6createEP11MDL_context13enum_mdl_type+0x26)[0x61a3e6]
/../mysqld(_ZN11MDL_context21try_acquire_lock_implEP11MDL_requestPP10MDL_ticket+0xa2)[0x61c4c2]
/../mysqld(_ZN11MDL_context12acquire_lockEP11MDL_requestm+0x63)[0x61cb53]

The keys parts of that, in reverse order are:

acquire_lock
try_acquire_lock_impl
enum_mdl_type

These 2 functions and variable, respectively, are only found in mdl.cc (and mdl.h) – the metadata lock files.

And after examining the MySQL source code, this stack trace tells us that there is an issue with the metadata locks. Specifically, it is trying to acquire a metadata lock (try_acquire_lock_impl), but it is failing (enum_mdl_type is set when try_acquire_lock_impl() fails, an assert is thrown, and mysqld crashes).

I’ve searched the bugs database for anything similar to this, but didn’t find any existing ones specific to try_acquire_lock_impl. That’s not to say there isn’t some yet-to-be-detected-bug here, but in this case, the machine was testing a custom application, so I strongly suspected the custom app causing the contention, which was the case.

So if you do happen to encounter this crash, then hopefully this will help you track down the problem.

Hope this helps, and happy troubleshooting. :)

 

Tags: , , , , , , , , , , ,
Posted in MySQL, SkySQL | 1 Comment »

Troubleshooting “Waiting for table metadata lock” Errors for both MyISAM and InnoDB Tables

March 14th, 2013

With the introduction of metadata locking in MySQL 5.5, it is much more common to see queries stuck in the “Waiting for table metadata lock” state.

If you see a query stuck in the “Waiting for table metadata lock” state, waiting for a MyISAM table, then it is because this table has been included as part of some uncommitted transaction (whether intentionally or not).

In fact, it could be as simple as issuing SET AUTOCOMMIT=0 followed by a SELECT against a MyISAM table (a transaction can be started 3 ways: START TRANSACTION, BEGIN, and SET AUTOCOMMIT=0). After that, the table will be locked to other DDL statements until the transaction is committed (or ended).

This metadata lock is meant for DDL statements, and so that is what it blocks. However, once a DDL statement is blocked and waiting, then all queries to that table (including SELECTs) are blocked. Here is how you can easliy reproduce it:

Connection #1:

create table t1 (id int) engine=myisam;
set @@autocommit=0;
select * from t1;

Connection #2:

alter table t1 rename to t2; <-- Hangs

You can see it hanging here (as well as a subsequent SELECT from a 3rd connection):

mysql> show full processlist;
+----+---+-------+---------------------------------+-----------------------------+
| Id |...| Time  | State                           | Info                        |
+----+---+-------+---------------------------------+-----------------------------+
|  1 |...|  1653 |                                 | NULL                        |
|  2 |...|     0 | NULL                            | show full processlist       |
|  3 |...| 18012 | Waiting for table metadata lock | alter table t1 rename to t2 |
|  4 |...| 17983 | Waiting for table metadata lock | select * from t1            |
+----+---+-------+---------------------------------+-----------------------------+

This is the expected behavior, but can be a little tricky to diagnose if it's happening to you, and you don't see any other queries currently running in SHOW FULL PROCESSLIST (and/or SHOW ENGINE INNODB STATUS). Furthermore, you cannot track this with any of the known lock tracking methods, such as InnoDB Lock Monitor, InnoDB Status, mysqladmin debug output, INFORMATION_SCHEMA, etc.), not to mention one generally doesn't associate transactions with MyISAM tables anyway.

You can also run into this same error in InnoDB, which one would not necessarily expect because of InnoDB's row-level locking.

However, when you introduce triggers, then it becomes easy to reproduce a similar situation.

create table t3 (id int) engine=innodb;
create table t4 (id int) engine=innodb;
delimiter |
CREATE TRIGGER t3_trigger AFTER INSERT ON t3
  FOR EACH ROW BEGIN
    INSERT INTO t4 SET id = NEW.id;
  END;
|
delimiter ;

Connection #1:

begin;
insert into t3 values (1);

Connection #2:

drop trigger if exists t3_trigger; <-- Hangs

So, if you see this with an InnoDB table, then check your SHOW ENGINE INNODB STATUS and look for prior transactions that hold locks and are still running. Those would be your likely culprits for what is blocking the DROP TRIGGER command (in this case).

Hope this helps and happy troubleshooting. :)

 

Tags: , , , , , , , , ,
Posted in MySQL, SkySQL | Comments Off

« Older Entries
Period Panties by Period Panteez Menstrual Underwear Menstruation PMS Panty