In any test that uses wait_all_purged.inc, ensure that InnoDB tables
will be created without persistent statistics.
This is a follow-up to commit cd04673a17
after a similar failure was observed in the innodb_zip.blob test.
The motivation of introducing the parameter
innodb_purge_rseg_truncate_frequency in
mysql/mysql-server@28bbd66ea5 and
mysql/mysql-server@8fc2120fed
seems to have been to avoid stalls due to freeing undo log pages
or truncating undo log tablespaces. In MariaDB Server,
innodb_undo_log_truncate=ON should be a much lighter operation
than in MySQL, because it will not involve any log checkpoint.
Another source of performance stalls should be
trx_purge_truncate_rseg_history(), which is shrinking the history list
by freeing the undo log pages whose undo records have been purged.
To alleviate that, we will introduce a purge_truncation_task that will
offload this from the purge_coordinator_task. In that way, the next
innodb_purge_batch_size pages may be parsed and purged while the pages
from the previous batch are being freed and the history list being shrunk.
The processing of innodb_undo_log_truncate=ON will still remain the
responsibility of the purge_coordinator_task.
purge_coordinator_state::count: Remove. We will ignore
innodb_purge_rseg_truncate_frequency, and act as if it had been
set to 1 (the maximum shrinking frequency).
purge_coordinator_state::do_purge(): Invoke an asynchronous task
purge_truncation_callback() to free the undo log pages.
purge_sys_t::iterator::free_history(): Free those undo log pages
that have been processed. This used to be a part of
trx_purge_truncate_history().
purge_sys_t::clone_end_view(): Take a new value of purge_sys.head
as a parameter, so that it will be updated while holding exclusive
purge_sys.latch. This is needed for race-free access to the field
in purge_truncation_callback().
Reviewed by: Vladislav Lesin
purge_sys_t::sees(): Wrapper for view.sees().
trx_purge_truncate_history(): Invoke purge_sys.sees() instead of
comparing to head.trx_no, to determine if undo pages can be safely freed.
The test innodb.cursor-restore-locking was adjusted by Vladislav Lesin,
as was the the debug instrumentation in row_purge_del_mark().
Reviewed by: Vladislav Lesin
This is joint work with Thirunarayanan Balathandayuthapani.
The MDL interface between InnoDB and the rest of the server
(in storage/innobase/dict/dict0dict.cc and in include/)
is my work, while most everything else is Thiru's.
The collection of InnoDB persistent statistics and the
defragmentation were not refactored to use MDL. They will
keep relying on lower-level interlocking with
fil_check_pending_operations().
The purge of transaction history and the background operations on
fulltext indexes will use MDL. We will revert
commit 2c4844c9e7
(MDEV-17813) because thanks to MDL, purge cannot conflict
with DDL operations anymore. For a similar reason, we will remove
the MDEV-16222 test case from gcol.innodb_virtual_debug_purge.
Purge is essentially replacing all use of the global dict_sys.latch
with MDL. Purge will skip the undo log records for tables whose names
start with #sql-ib or #sql2. Theoretically, such tables might
be renamed back to visible table names if TRUNCATE fails to
create a new table, or the final rename in ALTER TABLE...ALGORITHM=COPY
fails. In that case, purge could permanently leave some garbage
in the table. Such garbage will be tolerated; the table would not
be considered corrupted.
To avoid repeated MDL releases and acquisitions,
trx_purge_attach_undo_recs() will sort undo log records by table_id,
and purge_node_t will keep the MDL and table handle open for multiple
successive undo log records.
get_purge_table(): A new accessor, used during the purge of
history for indexed virtual columns. This interface should ideally
not exist at all.
thd_mdl_context(): Accessor of THD::mdl_context.
Wrapped in a new thd_mdl_service.
dict_get_db_name_len(): Define inline.
dict_acquire_mdl_shared(): Acquire explicit shared MDL on a table name
if needed.
dict_table_open_on_id(): Return MDL_ticket, if requested.
dict_table_close(): Release MDL ticket, if requested.
dict_fts_index_syncing(), dict_index_t::index_fts_syncing: Remove.
row_drop_table_for_mysql() no longer needs to check these, because
MDL guarantees that a fulltext index sync will not be in progress
while MDL_EXCLUSIVE is protecting a DDL operation.
dict_table_t::parse_name(): Parse the table name for acquiring MDL.
purge_node_t::undo_recs: Change the type to std::list<trx_purge_rec_t*>
(different container, and storing also roll_ptr).
purge_node_t: Add mdl_ticket, last_table_id, purge_thd, mdl_hold_recs
for acquiring MDL and for keeping the table open across multiple
undo log records.
purge_vcol_info_t, row_purge_store_vsec_cur(), row_purge_restore_vsec_cur():
Remove. We will acquire the MDL earlier.
purge_sys_t::heap: Added, for reading undo log records.
fts_sync_during_ddl(): Invoked during ALGORITHM=INPLACE operations
to ensure that fts_sync_table() will not conflict with MDL_EXCLUSIVE.
Uses fts_t::sync_message for bookkeeping.
Several race conditions between MDEV-15562 instant ALTER TABLE and purge
were observed.
The most obvious race condition resulted in a reported assertion failure in
dict_index_t::instant_add_field(): instant.n_core_fields == n_core_fields
would not hold if the table was emptied by purge after the time
dict_table_t::prepare_instant() was called.
During purge, it can turn out that the table is logically empty, only
containing a metadata record. If the metadata record is of the type
created by MDEV-11369 instant ADD COLUMN, it can be removed and
dict_index_t::clear_instant_add() can be called. This will convert the
table to the canonical non-instant format. (If the metadata record is
of the MDEV-15562 type, then it can only be deleted if the table becomes
empty as the result of rollback of an instant ALTER TABLE operation.)
row_purge_remove_clust_if_poss_low(): Add a debug check that ensures
that purge can never remove a MDEV-15562 metadata record.
ha_innobase::open(): Add a comment about the necessity of rolling
back any recovered instant ALTER TABLE transaction on the table.
instant_metadata_lock(): An auxiliary function to acquire a page latch
on the metadata record, to prevent race conditions.
dict_table_t::prepare_instant(), dict_index_t::instant_add_field(),
dict_table_t::rollback_instant(), innobase_instant_try():
Invoke instant_metadata_lock() in order to prevent race conditions.
dict_index_t::instant_add_field(): Correct debug assertions.
The == was guaranteed by code in dict_table_t::prepare_instant()
that was introduced in MDEV-15562. Due to the race condition,
we could occasionally have <=, but never >= like the code was
after MDEV-11369.
ha_innobase_inplace_ctx::instant_column(): Wrapper for
dict_table_t::instant_column(). Add debug assertions.
There was a race condition between ALTER TABLE and purge.
If a table turns out to be logically empty when instant ALTER TABLE
is executing, we will convert the table to the canonical format,
to avoid overhead during subsequent accesses, and to allow the
data file to be imported into older versions of MariaDB.
It could happen that at the time the table is logically empty,
there still exists an undo log record for updating the hidden
metadata record for an earlier instant ALTER TABLE operation.
If the table was converted to the canonical format before
purge processes this undo log record, the undo log record
could be referring to index fields that no longer exist,
causing a crash.
To prevent the race condition, we must delete the old undo log records.
We do this lazily by assigning a new table ID, so that the table lookup
for the old undo log records will fail.
dict_table_t::reassign_id(): Reassign the table_id to
effectively lazily delete old undo log records.
innobase_instant_try(): Invoke index->table->reassign_id() before
index->clear_instant_alter().
