With WolfSSL, the plugins is statically compiled, and enabled,
and defaults to autogenerating ssl keys, which was left unimplemented.
Thus, it spits out some [ERROR] on every startup.
Fixed by removing a couple some ifdefs. Allowed tcp_nossl to run on
Windows.
As WolfSSL is missing some APIs with FILE*, use related API that
accept BIO
, i.e
- BIO_new_file() instead of fopen()
- BIO_free instead of fclose()
- PEM_write_bio_PrivateKey() instead of PEM_write_PrivateKey()
- etc
A note about BIO and error reporting:
BIO_new_file sets the errno, therefore FILE_ERROR macro
produces good expected error messages, while SSL_ERROR unfortunately
creates something incomprehensible. Thus, FILE_ERROR is left in place
where it was used previously (fopen errors)
Curiously, removing APIs with FILE*, solves another bug MDEV-37343,
where server on Windows dies with obscure message as plugins tries to use
this function. OpenSSL_Applink supposed to be official solution against
such problems, but I could not get it to work properly, no matter how
much I tried. Avoiding APIs with FILE* in first place works best
set hashicorp_key_management_cache_version_timeout=60s by default
increase hashicorp_key_management_cache_timeout to 24h by default,
because key values should never change, but we don't want to remove
a variable for compatibility reasons
Commit 1d80e8e updated lock_operations to a rwlock from a mutex but didn't
update the PS instrumentation setup accordingly.
We update the PS setup accordingly so the lock is correctly instrumented in
performance_schema.rwlock_instances.
All new code of the whole pull request, including one or several files
that are either new files or modified ones, are contributed under the
BSD-new license. I am contributing on behalf of my employer Amazon Web
Services, Inc.
Nullability is decided in two stages-
1. Based on argument NULL-ness
Problem:
- COALESCE currently uses a generic logic- "Result of a function
is nullable if any of the arguments is nullable", which is wrong.
- IFNULL sets nullability using second argument alone, which incorrectly
sets the result to NULL even when first argument is not null.
Fix:
- Result of COALESCE and IFNULL is set to NULL only if all arguments are
NULL.
2. Based on type conversion safety of fallback value
Problem:
- The generic `Item_hybrid_func_fix_attributes` logic would mark the
function's result as nullable if any argument involved a type
conversion that could yield NULL.
Fix:
- For COALESCE and IFNULL, nullability is set to NOT NULL if the first
non-null argument can be safely converted to function's target return
type.
- For other functions, if any argument's conversion to target type could
result in NULL, the function is marked nullable.
Tests included in `mysql-test/main/func_hybrid_type.test`
Fix AWS SDK build, it has changed substantionally since the plugin was
introduced. There is now a bunch of intermediate C libraries, aws-cpp-crt
and others, and for static linking, the link dependency must be declared.
Also support AWS C++ SDK in vcpkg package manager.
Limit size of server_audit_file_path value
Currently, the length of this value is not checked and can cause a buffer
overflow if given a long file path specifying a directory.
In file_logger:logger_open(), there is a check:
```
if (new_log.path_len+n_dig(rotations)+1 > FN_REFLEN)
// handle error
```
As n_dig(rotations) may return up to 3, this inherently limits the file path to
FN_REFLEN - 4 characters.
All new code of the whole pull request, including one or several files that are
either new files or modified ones, are contributed under the BSD-new license. I
am contributing on behalf of my employer Amazon Web Services, Inc.
Disallow range optimization for BETWEEN when casting one of the arguments
from STRING to a numeric type would be required to construct a range for
the query.
Adds a new method on Item_func_between called can_optimize_range_const
which allows range optimization when the types of the arguments to BETWEEN
would permit it.
The problem is that copy function was used in field list but never
copied in this execution path.
So copy should be performed before returning result.
Protection against uninitialized copy usage added.
through pointer to incorrect function type.
Redoing a new plugin interface for an obsolete protocol was
too much so we just remove the UBSAN testing on the function.
It wasn't possible to just disable funtion-type-mismatch
in _ma_unique_hash, skip_trailing_space, my_hash_sort_mb_nopad_bin and my_strnncollsp_utf8mb4_bin
UBSAN detected the nullptr-with-offset in a few places
when handling empty blobs.
Fix:
- Adding DBUG_ASSERT(source_string) into all hash_sort() implementations
to catch this problem in non-UBSAN debug builds.
- Fixing mi_unique_hash(), mi_unique_comp(),
_ma_unique_hash(), _ma_unique_comp() to replace NULL pointer to
an empty string ponter..
Note, we should also add DBUG_ASSERT(source_string != NULL) into
all implementations of strnncoll*(). But I'm afraid the patch
is going to be too long and too dangerous for 10.5.
UUID::cmp() correctly compared:
- two swapped v1 UUIDs
- two non-swapped v6 UIDs
but v1 vs v6 were not compared correctly.
Adding a new method cmp_swap_noswap() and using
it in UUID::cmp() to compare two value of different swapness.
