MySQL Bugs: #2: Does not make Toast

Bug #2	Does not make Toast
Submitted:	12 Sep 2002 5:26	Modified:	26 Aug 2021 14:23
Reporter:	Mark Matthews	Email Updates:
Status:	Closed	Impact on me:	None
Category:	Connector / J	Severity:	S7 (Test Cases)
Version:	3.0.0	OS:	Linux (Windows and Others))
Assigned to:		CPU Architecture:	Other

Description:
Other JDBC drivers I have used make toast for breakfast.

MySQL Connector/J doesn't make toast, it can only pour a bowl of froot loops.

When I ask it to make toast, the "NoToast" exception is thrown. I think you should look into this.

How to repeat:

Thank you for taking the time to write to us, but this is not
a bug. Please double-check the documentation available at
http://www.mysql.com/documentation/ and the instructions on
how to report a bug at http://bugs.mysql.com/how-to-report.php

This is as expected. MySQL Connector/J does not make toast. I don\'t think it makes froot loops either (will have to check).

I believe you should look into a device called a \"toaster\" to make your toast.

Of course if you want to sponsor the \"toast\" feature, we would be glad to implement it for you.

my PS2 makes toast ... it even looks like a toaster.

;-D

This report is almost 10 years old now and C/J *still* doesn't make toast? Sheesh.

i thought it's a server bug.  if the server exported a menu we could wish for breakfast in bed.

File upload test

Attachment: test_file.txt (text/plain), 21 bytes.

Happy Birthday Bug!

11 years now without toast ... happy birthday bug#2!

the best bug ever.

Added the following entry to the Connector/J version who-knows-what for this
historical Bug#2, which was a test and was supposed to be ignored:

"Connector/J did not make toast. This was due to the missing of the Toaster
class, which this fix has added. However, the class is only accessible when a
user is extremely hungry. If Connector/J still does not make toasts for you,
it is only because you are not hungry enough."

Noted in the docs in response to customer issue:

"The Toaster class must be initialised by calling plug_in(). Otherwise, instances can be created and their toast() methods invoked as normal, but getColour() in such cases continues to return CJ_TOAST_CLR_WHITE without throwing any errors."

This update has been committed to our documentation repository and should appear in the published docs online very soon.

This actually has been solved finally!
https://www.percona.com/blog/2015/11/25/smallest-mysql-server-world-60-can-even-make-toast...

I tried to make a toast using a stored procedure but, as I mentioned several time, current MySQL implementation of SP is too limited and slow. I worked hard for a month, and now, even when the procedure doesn't fail, it takes several hours to make a complete toast. What's worst is that there is no way to return it to the caller (just like cursors), so it remains in memory and is eventually deleted.

I hope to see improvements in a future MySQL version.

I've created a patch for this bug here: https://github.com/alikrubin/lib_mysqludf_sys
(also attached 2 files).
Full description for the patch is here: https://www.percona.com/blog/2016/04/01/fixing-mysql-bug2-now-mysql-makes-toast/

New regression in 8.0.1

#0  0x00007f11bf9cc741 in __pthread_kill (threadid=<optimized out>, signo=6) at ../nptl/sysdeps/unix/sysv/linux/pthread_kill.c:61
#1  0x00000000018514ca in my_write_core (sig=6) at /git/mysql-8.0.1_dbg/mysys/stacktrace.c:249
#2  0x0000000000e81e91 in handle_fatal_signal (sig=6) at /git/mysql-8.0.1_dbg/sql/signal_handler.cc:223
#3  <signal handler called>
#4  0x00007f11bdd601d7 in __GI_raise (sig=sig@entry=6) at ../nptl/sysdeps/unix/sysv/linux/raise.c:56
#5  0x00007f11bdd618c8 in __GI_abort () at abort.c:90
#6  0x0000000001bcd31c in ut_dbg_assertion_failed (expr=0x2233ba3 "purge_sys->n_stop == 0", file=0x2232cb8 "/git/mysql-8.0.1_dbg/storage/innobase/srv/toast.cc", line=3070)
    at /git/mysql-8.0.1_dbg/storage/innobase/ut/ut0dbg.cc:67
#7  0x0000000001b63d44 in check_if_burned (sandwich=0x7f11b1be8868,toaster_slot=13) at /git/mysql-8.0.1_dbg/storage/innobase/srv/maketoast.cc:3070
#8  0x0000000001b640d7 in make_toast (arg=0x0) at /git/mysql-8.0.1_dbg/storage/innobase/srv/maketoast.cc:1000
#9  0x00007f11bf9c7dc5 in start_thread (arg=0x7f118b7fd700) at pthread_create.c:308
#10 0x00007f11bde2273d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:113

Connector/J is now throwing AppleJacksException on OS X.   Still no toast :(

personal emoji for the bug 🍞

Happy birthday bug #2!

Fix confirmed by @kk2170 in Japan,
with MySQL 8.0.13 using Arduino and arudino_filmata Ruby library;
 https://gihyo.jp/dev/serial/01/mysql-road-construction-news/0087
(an article in the Japanese IT media Gihyo.jp, written in Japanese)

I would argue that making toasts would break JDBC specification.

It would also sensibly reduce salability, because making a toast acquires an X lock on the device for about one minute.

Last but not least, this will break backward compatibility. Consider all those apps that would start to make a toast every time a page is visited. Obesity would become a huge problem for DBAs.

This is a test of comments copying

Test comment

testing

s/testing/toasting

hello

also testing

Posted by developer:
 
Test

Posted by developer:
 
test

Posted by developer:
 
Test after restart

can we make a toast by mysql ... now..?

Posted by developer:
 
Testing C1

Posted by developer:
 
Test C3S

Posted by developer:
 
Test C4S2

Posted by developer:
 
Test C5

Posted by developer:
 
test C6S

Posted by developer:
 
test C 11

Posted by developer:
 
Test C13

Posted by developer:
 
Can't repro, makes toast just fine here.

Posted by developer:
 
ping

Posted by developer:
 
Test C27

Posted by developer:
 
Test C28

Posted by developer:
 
Test C30

Posted by developer:
 
Test C30

Posted by developer:
 
Test C31(b)

Posted by developer:
 
Test C32

Posted by developer:
 
Test C40

Posted by developer:
 
Test C42

Posted by developer:
 
Another test

Posted by developer:
 
Another test

It's been 19 years. I'm starting to be afraid the developers don't appreciate how urgent this is.

Posted by developer:
 
Testing 628

Posted by developer:
 
testing - c ignore

Posted by developer:
 
Test w/comment