Sunday, May 31, 2020

Dynamic Tracing of Memory Allocations in MySQL: First Steps with perf

I often have to deal with customers claiming there is a memory leak in MySQL affecting their production servers. To prove the leak is real and show where it happens usually running with tcmalloc and relying to it's heap profiler, or even more intrusive, running under Valgrind Massif is needed. This is normal when the test case demonstrating the leak is isolated and we prove a leak for some bug report. But hardly anyone in production in a normal situation would agree to do this, as performance impact is very serious.

So, recently I decided to check if dynamic tracing (along the lines of this page by Brendan Gregg, with Flame Graphs to show suspicious code path) is a more reasonable alternative. As I have to work with different Linux versions, often as old as CentOS 6, I can not rely on bcc tools for this. I mentioned more than once that in internal discussion that adding dynamic probe for malloc() (and other memory allocating functions, if you prefer) with perf may already be a good start and give some data to study. So, time to try this approach myself (so that I am ready to suggest it to customers with all the details at hand) and continue my series of posts on dynamic tracing. It can not be all the time about bugs, readers needs solutions as well :)

For this post I decided to work with latest and greatest MySQL 8.0.20 that I have built from source on my good old almost irrelevant Ubuntu 16.04 netbook:
openxs@ao756:~$ uname -a
Linux ao756 4.4.0-178-generic #208-Ubuntu SMP Sun Apr 5 23:45:10 UTC 2020 x86_64 x86_64 x86_64 GNU/Linux
openxs@ao756:~$ cd dbs/8.0
openxs@ao756:~/dbs/8.0$ bin/mysql -uroot --socket=/tmp/mysql8.sock -e "show variables like '%version%'"
| Variable_name            | Value                 |
| immediate_server_version | 999999                |
| innodb_version           | 8.0.20                |
| original_server_version  | 999999                |
| protocol_version         | 10                    |
| slave_type_conversions   |                       |
| tls_version              | TLSv1,TLSv1.1,TLSv1.2 |
| version                  | 8.0.20                |
| version_comment          | Source distribution   |
| version_compile_machine  | x86_64                |
| version_compile_os       | Linux                 |
| version_compile_zlib     | 1.2.11                |
The reason is simple. Looks like I have more readers while writing about MySQL, not some fork of it :) I hope this tend will change soon.

First of all, I have to check what memory allocation library is used (the details of adding probe and even some features available to probe may be different for jemalloc, for example):
openxs@ao756:~/dbs/8.0$ ldd bin/mysqld =>  (0x00007ffe8afb6000) => /lib/x86_64-linux-gnu/ (0x00007fc9bdfe2000) => /home/openxs/dbs/8.0/bin/../lib/private/ (0x00007fc9bdd8d000) => /lib/x86_64-linux-gnu/ (0x00007fc9bdb85000) => /lib/x86_64-linux-gnu/ (0x00007fc9bd91c000) => /lib/x86_64-linux-gnu/ (0x00007fc9bd4d7000) => /lib/x86_64-linux-gnu/ (0x00007fc9bd2d3000) => /lib/x86_64-linux-gnu/ (0x00007fc9bd0d1000) => /usr/lib/x86_64-linux-gnu/ (0x00007fc9bcec6000) => /usr/lib/x86_64-linux-gnu/ (0x00007fc9bcb44000) => /lib/x86_64-linux-gnu/ (0x00007fc9bc83b000) => /lib/x86_64-linux-gnu/ (0x00007fc9bc625000) => /lib/x86_64-linux-gnu/ (0x00007fc9bc25b000)
        /lib64/ (0x00007fc9be1ff000)
I do not see tcmalloc or jemalloc in the output above, but libc is linked dynamically. So, malloc() implementation would likely come from it (/lib/x86_64-linux-gnu/ binary, to be specific). Let me add the probe (see this older post for more details, some details about probes syntax presented here are also useful):
openxs@ao756:~$ sudo perf probe -x /lib/x86_64-linux-gnu/ 'malloc size=%di:s64'
[sudo] password for openxs:
Added new events:
  probe_libc:malloc    (on malloc in /lib/x86_64-linux-gnu/ with size=%di:s64)
  probe_libc:malloc_1  (on malloc in /lib/x86_64-linux-gnu/ with size=%di:s64)

You can now use it in all perf tools, such as:

        perf record -e probe_libc:malloc_1 -aR sleep 1

One of the first things I want to be able to find out is how many bytes are allocated by malloc(). In case of libc malloc() on this Ubuntu I could just use the fist and only argument name (you'd be surprised what it really is, try to find out!) or just $params (see man perf-probe for details).  But in a general case (like with jemalloc on the same system) I should better refer to it based on the way it is passed - via %DI (not short form, NOT %EAI) register on x86 systems. This is why you see size=%di:s64 in the probe definition. Note also that to get real value as integer I had to use s64 type (for signed 64-bit integer, even though I;d expect the argument to be unsigned). By default and as unsigned we get hex value, and one day I'd like to be able to count also bytes allocated, not just calls.

I wanted to add return probe for the same function, in a hope to store (one day) the pointer and then maybe check (with other probe) if it is ever explicitly freed. But in my case it is not possible to get the return value:
openxs@ao756:~$ sudo perf probe -x  /lib/x86_64-linux-gnu/ malloc_ret='malloc%return $retval'
Failed to find "__libc_malloc%return",
 because __libc_malloc is an inlined function and has no return point.
Added new event:
  probe_libc:malloc_ret (on malloc%return in /lib/x86_64-linux-gnu/ with $retval)

You can now use it in all perf tools, such as:

        perf record -e probe_libc:malloc_ret -aR sleep 1
I am yet to check the reason, but according to the above the function is inlined and thus does not have a return point. Sounds weird. Note that for jemalloc I was able to add similar probe successfully and get the pointer value traced.

So, what do I get when I try to trace this probe in perf? Let me show. If I run some stupid test surely allocating memory in one shell:
openxs@ao756:~/dbs/8.0$ bin/mysqlslap -uroot --socket=/tmp/mysql8.sock --concurrency=12 --create-schema=test --no-drop --number-of_queries=300000 --iterations=2 --query='select @a := repeat("a", 1000);'
While in the other I'll try to record some samples:
openxs@ao756:~$ sudo perf record -e 'probe_libc:malloc' -aRg sleep 10
[sudo] password for openxs:
Lowering default frequency rate to 3250.
Please consider tweaking /proc/sys/kernel/perf_event_max_sample_rate.
[ perf record: Woken up 710 times to write data ]
[ perf record: Captured and wrote 179.217 MB (1017163 samples) ]
Then I'll get the output like this from perf script:
openxs@ao756:~$ sudo perf script | grep mysqld | more
mysqld   632 [001] 218853.089892: probe_libc:malloc: (7fc7312b7130) size=120
                  a18de2 std::__detail::_Hashtable_alloc<Malloc_allocator<std::_
_detail::_Hash_node<TABLE_LIST*, true> > >::_M_allocate_buckets (/home/openxs/db
                  a19061 lock_table_names (/home/openxs/dbs/8.0/bin/mysqld)
                  a1adc8 open_tables (/home/openxs/dbs/8.0/bin/mysqld)
                  a1b576 open_tables_for_query (/home/openxs/dbs/8.0/bin/mysqld)
                  ae7313 Sql_cmd_dml::prepare (/home/openxs/dbs/8.0/bin/mysqld)
                  aef86f Sql_cmd_dml::execute (/home/openxs/dbs/8.0/bin/mysqld)
                  a92b0a mysql_execute_command (/home/openxs/dbs/8.0/bin/mysqld)
                  a97824 mysql_parse (/home/openxs/dbs/8.0/bin/mysqld)
                  a99605 dispatch_command (/home/openxs/dbs/8.0/bin/mysqld)
                  a9abd0 do_command (/home/openxs/dbs/8.0/bin/mysqld)
                  bba6b0 handle_connection (/home/openxs/dbs/8.0/bin/mysqld)
                 20d0355 pfs_spawn_thread (/home/openxs/dbs/8.0/bin/mysqld)
mysqld  2795 [000] 218853.091016: probe_libc:malloc: (7fc7312b7130) size=1040                  bb7ed3 String::real_alloc (/home/openxs/dbs/8.0/bin/mysqld)
                  dda12f Item_func_repeat::val_str (/home/openxs/dbs/8.0/bin/mys
                  da5272 Item_func_set_user_var::check (/home/openxs/dbs/8.0/bin
                  da5874 Item_func_set_user_var::val_str (/home/openxs/dbs/8.0/b
                  d1d1cf Item::send (/home/openxs/dbs/8.0/bin/mysqld)
                  a23d29 THD::send_result_set_row (/home/openxs/dbs/8.0/bin/mysq
                  eb86c1 Query_result_send::send_data (/home/openxs/dbs/8.0/bin/
                  b55524 SELECT_LEX_UNIT::ExecuteIteratorQuery (/home/openxs/dbs
                  b5566e SELECT_LEX_UNIT::execute (/home/openxs/dbs/8.0/bin/mysq
                  ae8f52 Sql_cmd_dml::execute_inner (/home/openxs/dbs/8.0/bin/my
                  aefc37 Sql_cmd_dml::execute (/home/openxs/dbs/8.0/bin/mysqld)
                  a92b0a mysql_execute_command (/home/openxs/dbs/8.0/bin/mysqld)
                  a97824 mysql_parse (/home/openxs/dbs/8.0/bin/mysqld)
                  a99605 dispatch_command (/home/openxs/dbs/8.0/bin/mysqld)
Note the following:
  • I tried to trace all calls to malloc() from glibc, for all processes, with stack traces. As a result I've got 179 MB of raw data in over just 10 seconds, with default frequency of sampling! This is too much for any pracrtical purpose, so make sure to set smaller frequency (like -F99) or performance impact may be too seriois. With newer kernels and eBPF this impact may be notably reduced.
  • I've got both allocation size reported (and may get a good use for it one day), and stack traces, showing where the allocation is coming from! This is exactly what developers need to check if it is really a possible leak.
Now what I can I do with the above? I can use perf report (sudo perf report > /tmp/report.txt) and get the call graphs in text:
openxs@ao756:~$ cat /tmp/report.txt | more                                     
# To display the header info, please use --header/--header-only options.
# Total Lost Samples: 0
# Samples: 556  of event 'probe_libc:malloc'
# Event count (approx.): 4935574
# Children      Self  Command          Shared Object        Symbol

# ........  ........  ...............  ...................  ....................
   87.56%    87.56%  mysqlslap         [.] malloc
                     |--86.77%-- mysql_real_query
                     |          cli_read_query_result
                     |          read_com_query_metadata
                     |          cli_read_metadata_ex
                     |          MEM_ROOT::AllocSlow
                     |          MEM_ROOT::AllocBlock
                     |          malloc
                      --0.79%-- mysql_store_result
    10.18%     0.00%  mysqld           mysqld               [.] do_command
                        |--9.00%-- mysql_parse
                        |          |
                        |          |--7.88%-- mysql_execute_command
                        |          |          Sql_cmd_dml::execute
                        |          |          |
                        |          |          |--6.64%-- Sql_cmd_dml::prepare
                        |          |          |          open_tables_for_query
                        |          |          |          open_tables
                        |          |          |          lock_table_names
                        |          |          |          |
                        |          |          |          |--6.04%-- std::__detai
l::_Hashtable_alloc<Malloc_allocator<std::__detail::_Hash_node<TABLE_LIST*, true
> > >::_M_allocate_buckets
                        |          |          |          |          malloc
                        |          |          |          |
                        |          |          |           --0.59%-- get_and_lock
                        |          |          |                     std::__detai
ic_string<char, std::char_traits<char>, std::allocator<char> >, true> > >::_M_al
                        |          |          |                     malloc
                        |          |          |
                        |          |           --1.24%-- Sql_cmd_dml::execute_inner
                        |          |                     SELECT_LEX_UNIT::execute
                        |          |                     SELECT_LEX_UNIT::ExecuteIteratorQuery
                        |          |                     Query_result_send::send_data
                        |          |                     THD::send_result_set_row
                        |          |                     Item::send
                        |          |                     |
                        |          |                     |--0.66%-- Item_func_set_user_var::val_str
                        |          |                     |          Item_func_set_user_var::check
                        |          |                     |          Item_func_repeat::val_str
                        |          |                     |          String::real_alloc
                        |          |                     |          malloc
                        |          |                     |
                        |          |                      --0.58%-- user_var_entry::val_str
                        |          |                                String::copy
                        |          |                                String::real_alloc
                        |          |                                malloc
                        |          |
                        |           --1.12%-- parse_sql
                        |                     |
                        |                     |--0.66%-- THD::sql_parser
                        |                     |          MYSQLparse
                        |                     |          THD::raise_condition
                        |                     |          Diagnostics_area::push_warning
                        |                     |          MEM_ROOT::AllocSlow
                        |                     |          MEM_ROOT::AllocBlock
                        |                     |          malloc
                        |                     |
                        |                      --0.46%-- Diagnostics_area::copy_sql_conditions_from_da
                        |                                Diagnostics_area::push_warning
                        |                                Diagnostics_area::push_warning
                        |                                MEM_ROOT::AllocSlow
                        |                                MEM_ROOT::AllocBlock
                        |                                malloc
                         --1.18%-- THD::swap_rewritten_query
I can also create a flame graph of number of malloc() calls per function/sequence of calls. I had to modify steps from the original source a big, to come up with the following:
openxs@ao756:~$ sudo perf script > out.stack
openxs@ao756:~$ git/FlameGraph/ < out.stack | git/FlameGraph/ --color=mem --title='malloc( Fale Graph' --countname="calls" -- > malloc.svg
For this test I've included allocations for all the processes and had to use script, as stack traces are coming in perf format.

The result (actually a part of it, just for the mysqld process, copy/pasted as .png static file) may look as follows:

Sample malloc() calls flame graph build from one of tests.
Now you can see where most malloc() calls happen, and this is the first step in checking where leaks may come from if we really suspect them. Quite a lot of fun for a simple probe added with perf, on any Linux system younger than 10 years!

Next time I am going to apply this approach to some real life use case that looks like a leak and see if anything useful, not just fun, can be derived from it. For now I know it's doable and I am sure about the steps (documented here and more in my background notes) to trace malloc() with perf dynamic probe.

Saturday, May 30, 2020

Fun with Bugs #99 - On MySQL Bug Reports I am Subscribed to, Part XXXIII

In my previous post in this series I've commented on some interesting MySQL bug reports that were added during the second half of April. Time to move on to bugs reported in May, 2020, as we are quickly approaching MySQL Bug #100000 soon and I want to write a separate post for this great achievement :)

Here is the list:
  • Bug #99432 - "Improving memory barrier during rseg allocation". Nice contribution by my former colleague in Percona, Krunal Bauskar, who now works on making MySQL better for ARM processors. According to his findings, the use of a relaxed memory model improves performance on ARM by up to 2%. See also yet another bug report with a contribution that matters for ARM, Bug #99556 - "Avoid sequentially consistent atomics for atomic counters" (contributed by Sergey Glushchenko from Percona).
  • Bug #99444 - "New HASH JOIN order problem". One should not expect and rely on any specific order unless explicit ORDER BY is used, so formally this report by Gabor Aron is "Not a Bug". I put it into this list as several other community members helped him a lot in understanding why results with HASH_JOIN optimization in newer versions are still valid and what are the ways to get the results with the desired ordering efficiently. Guilhem Bichot, for instance, suggested two different ways, using window function and lateral table. Useful reading in any case!
  • Bug #99458 - "i_s_fts_index_cache_fill_one_index() is not protect by the lock". Looks like even crashes are possible as a result, based on comments. Nice finding by Haiqing Sun.
  • Bug #99459 - "SQL run with GROUP_MIN_MAX may infinite loop and never return". After some discussion around the validity and severity of bug reports where test case involved adding DEBUG_SYNC() to show the problem in a predictable way, this great bug report by Ze Yang was verified. All MySQL GA versions are affected, including 8.0.20! As a side note, I'd prefer NOT to read such discussions any more. They are wasting time of all parties involved.
  • Bug #99499 - "Incorrect result when constant equailty expression is used in LEFT JOIN condition". This bug that affects MySQL 5.7.x only (it was fixed in MySQL 8.0.17+ and in 5.6 code was different) was reported by Marcos Albe from Percona. 
  • Bug #99504 - "Generated column incorrect on INSERT when based on column w/ expression DEFAULT". Several problems are highlighted in the complex enough test case submitted by Brad Lanier.
  • Bug #99582 - "Reduce logging of new doublewrite buffer initialization which is confusing". 180 lines or so are added when --log-error-verbosity is set to 3. As a workaround one can add:
    to the [mysqld] section of the .cnf file. This problem was reported by Simon Mudd. Make sure to read all comments.
  • Bug #99591 - "Option --tc-heuristic-recover documentation wrong, missing details". In reality it does not work with more than one XA-capable engine installed. I wish fine manual documents the reality, not the good intentions of the past. This documentation request was added by Sami Ahlroos.
  • Bug #99593 - "Performance issues in 8.0.20". It seems to be yet another TempTable engine problem that caused regression comparing to MySQL 5.7. At least this:

    • SET GLOBAL internal_tmp_mem_storage_engine=MEMORY;

    is a workaround. The bug (a duplicate of internal Bug #30562964) was reported by billy noah and is fixed in upcoming MySQL 8.0.21.
  • Bug #99601 - "Broken Performance using EXIST function, increasing execution time each loop". This regression bug (without tag, but who cares...) in MySQL 8.0 was reported by Ronny Görner and minimal test case demonstrating that the problem is actually with function call was contributed by Shane Bester.
  • Bug #99643 - "innobase_commit_by_xid/innobase_rollback_by_xid is not thread safe". This bug was reported by Zhai Weixiang, who had also suggested the fix in the code.
  • Bug #99717 - "Performance regression of parallel count". Great bug report with code analysis and ready to use MTR test case from Ze Yang. Sunny Bains already confirmed that the problematic code is going to be removed.

To summarize:
  1. I am happy to see Oracle engineers explaining community bug reporters the reasons and possible solutions for the problems they hit that are not actually caused by any bug in MySQL. I tried to do this as well, whenever possible, while working on MySQL bugs...
  2. We can still find speculations that if the bug is repeatable only by adding DEBUG_SYNC() or similar debug lines, then it can not be verified or gets lower severity... IMHO this is nonsense, as there are many high severity verified real bug reports where this method is used to demonstrate the problem clearly. Just stop it!

Sunday, May 24, 2020

Fun with Bugs #98 - On MySQL Bug Reports I am Subscribed to, Part XXXII

There are many things to write about MySQL this week. It turned 25 years old, to begin with! Quite successful the first ever Percona Live ONLINE 24 hour conference also happened this week, and I've presented my talk there...

But this is a blog of former MySQL Entomologist, so when I have nothing ready to share about something exciting or immediately useful, I write about MySQL bugs. This is the case today as well. I need more time to think about MySQL history and write down the details to complement my presentation etc. Previous post in this series appeared a month ago, so let me present a yet another list of InnoDB, optimizer, replication and few other bugs reported by MySQL Community users since April 18, 2020 and before May, 2020:
  • Bug #99295 - "InnoDB in-memory estimate is wrong after an index is created". As Øystein Grøvlen found out, even though the entire newly added covering index is in the buffer pool, the buf_stat_per_index_t::get() function in MySQL 8 estimates just 1 page is in memory, and as a result the index is not used.
  • Bug #99326 - "undo truncation might still not be crash safe". After nice explanations of what may happen from Kevin Lewis:
    "After and internal discussion with Sunny Bains, I think I understand the concern better. Let's assume that a redo log is so large that it contains redo entries for all 512 Space IDs of an undo tablespace that is being truncated too often. In other words, even though each truncate removes old pages from the buffer pool and flushes newly created pages, it does not actually cause a checkpoint for each truncation like it did in 5.7. So the redo log can possibly contain records for more than 512 space IDs.

    There is a worklog tested and pushed to the 8.0.21 release branch that fixes this highly unlikely possibility.

    As part of WL#11819, we keep a count of the number of truncations that have happened between checkpoints. So if there is more than (512 / 8) truncations between checkpoints, then no more truncations can happen on that undo space until the next checkpoint happens."
    this (regression vs 5.7) bug report by Zanye Zjy was closed as "Not a Bug". This is a totally wrong status and handling for such a case. 
  • Bug #99354 - "Nondeterministic stored function returns incorrect results". This interesting bug was reported by Jacob Chafik. If anyone cares, it is still repeatable on 8.0.20 (and not repeatable on MariaDB 10.4.x):
    mysql> SELECT IF(COUNT( > 0, "Success", "Failure") "Result" FROM t1 WHERE 1 IN (SELECT FROM t1 t1_inner INNER JOIN t2 WHERE f1( > 0);
    | Result  |
    | Failure |
    1 row in set (0,04 sec)

    mysql> SELECT IF(COUNT( > 0, "Success", "Failure") "Result" FROM t1 WHERE 1 IN (SELECT FROM t1 t1_inner INNER JOIN t2 WHERE > 0);
    | Result  |
    | Success |
    1 row in set (0,00 sec)

    mysql> explain SELECT IF(COUNT( > 0, "Success", "Failure") "Result" FROM t
    1 WHERE 1 IN (SELECT FROM t1 t1_inner INNER JOIN t2 WHERE f1(
    > 0);

    | id | select_type | table    | partitions | type | possible_keys | key  | key_len | ref  | rows | filtered | Extra                                      |
    |  1 | SIMPLE      | t1_inner | NULL       | ALL  | NULL          | NULL | NULL    | NULL |    2 |    50.00 | Using where                                |
    |  1 | SIMPLE      | t2       | NULL       | ALL  | NULL          | NULL | NULL    | NULL |    2 |   100.00 | FirstMatch                                 |
    |  1 | SIMPLE      | t1       | NULL       | ALL  | NULL          | NULL | NULL    | NULL |    2 |   100.00 | Using where; Using join buffer (hash join) |
    3 rows in set, 1 warning (0,00 sec)

    mysql> explain SELECT IF(COUNT( > 0, "Success", "Failure") "Result" FROM t1 WHERE 1 IN (SELECT FROM t1 t1_inner INNER JOIN t2 WHERE > 0);
    | id | select_type | table    | partitions | type | possible_keys | key  | key_len | ref  | rows | filtered | Extra                         |
    |  1 | SIMPLE      | t1_inner | NULL       | ALL  | NULL          | NULL | NULL    | NULL |    2 |    50.00 | Using where                   |
    |  1 | SIMPLE      | t2       | NULL       | ALL  | NULL          | NULL | NULL    | NULL |    2 |    50.00 | Using where; FirstMatch       |
    |  1 | SIMPLE      | t1       | NULL       | ALL  | NULL          | NULL | NULL    | NULL |    2 |   100.00 | Using join buffer (hash join) |
    3 rows in set, 1 warning (0,00 sec)

    mysql> select version();
    | version() |
    | 8.0.20    |
    1 row in set (0,00 sec)
    Wrong results should not be produced for logically equivalent queries, even if plans are different. Unfortunately this regression (since 5.7.5+) bug had not got the "regression" tag.
  • Bug #99359 - "Order by in group_concat with prepare statement returns unexpected results". Two executions of the same statement with the user variable having the same value assigned should either both work or fail with error message, it's a matter of consistency. This is not the case in a situation described in the bug report from Feng Liyuan.
  • Bug #99363 - "Innodb_data_pending_fsyncs shows incorrect value". In this bug report SeongUck Lee clearly shown that the problem happens and gave some hints based on source code review what could cause this regression in 8.0 comparing to 5.7. Moreover, after adding debug assertion he managed to show how it is hit and what wrong values are produced, in gdb. Still this report for now ended up as "Not a Bug" (until a test case is presented, reportedly. What a shame!
  • Bug #99377 - "Assertion `thd->get_transaction()->is_empty(Transaction_ctx::STMT)' failed in ". Surely dropping and re-creating the mysql.general_log table is a corner case, but even in debug builds it should end up with some proper error message, not just assertion failure. This bug was reported by Roel Van de Paar.
  • Bug #99381 - "ORDER_BY Index-level optimizer hint implies NO_JOIN_INDEX for second table". The bug reporter, Jesper Wisborg Krogh, my former colleague in MySQL Support, knows a lot about MySQL queries optimization. Make sure to read his last book, "MySQL 8 Query Performance Tuning: A Systematic Method for Improving Execution Speeds". This time he found a problem with hints.
  • Bug #99398 - "Data in wrong row on left join". This regression bug in MySQL 8.0.20 reported by Soner Sayakci is already fixed in upcoming MySQL 8.0.21. Good news.
  • Bug #99412 - "Threads_running becomes scalability bottleneck on multi-node NUMA topologies". Nice bug report from Sergey Glushchenko, with a patch contributed. perf annotate was used to demonstrate the problem. Good to know Percona engineers keep working to make MySQL perform better!
  • Bug #99413 - "Constant propagation get the wrong result when mix with the different collations". Again, this bug report from Wj Huang was closed as 'Not a Bug" based mostly on the results from 8.0.21+ that is not yet released. This is either a "Duplicate" if the exact fix that helped in 8.0.21+ can be isolated, or "Can't repeat", but setting it to "Not a Bug" seems wrong to me.
  • Bug #99414 - "Where exists performance 8.0.16 onwards". This optimizer regression bug in 8.0.16+ (comparing to 5.7) was reported by Jonathan Taylor. See also a comment by Øystein Grøvlen about the tool available at to pre-process and get a condensed trace for the join optimizer, that helps to understand what's going on.
Let me stop for now. Many more interesting bugs were reported in May, so stay tuned!

Happy Birthday, MySQL and Sakila!
To summarize:
  1.  Now (regression!) bugs are sometimes closed as "Not a Bug" after explanation what the problem is and statement that there is a fix in a version not yet released. Not as a "Duplicate", but "Not a Bug", after confirming the problem. This is a new level of wrong bugs handling!
  2. Same with a bug where bug reporter is still working on a repeatable test case, after showing the problem happened with some evidence, "Not a Bug".
  3. Oracle does a good job recently in making sure MySQL Server bugs are not hanging around without reaction and are processed fast, but looks like this is partially achieved by lowering standards of bugs processing that were established over last 15+ years. Closing reports by all means as soon as possible as "Not a Bug" is not a way to go. This is sad.