Hangfire & MariaDB locking issue

Question

There is a software hangfire that uses mysql/mariadb as the backend. When the hangfire database was dropped, some InnoDB tables were left undropped and locked in some unusual way.

• hangfire_hash
• hangfire_job
• hangfire_jobqueue
• hangfire_state

Any DDL/DML like DROP TABLE, TRUNCATE, ALTER TABLE, DELETE or UPDATE causes an error:

ERROR 1205 (HY000): Lock wait timeout exceeded; try restarting transaction

On the other hand, SELECT * FROM hangfire_job; or SELECT count(*) FROM hangfire_job; works as expected. I had found an article with exactly the same issue and, most surprisingly, locked and undroppable tables are also the same: https://stackoverflow.com/q/78422902/1200574

What could be causing this strange behavior and is there a way to deal with these immortal tables?

UPD.1

MariaDB [sys]>  SELECT * FROM sys.innodb_lock_waits \G
*************************** 1. row ***************************
                wait_started: 2024-12-18 20:33:03
                    wait_age: 00:00:23
               wait_age_secs: 23
                locked_table: `emails`.`hangfire_jobqueue`
                locked_index: NULL
                 locked_type: TABLE
              waiting_trx_id: 7691700269
         waiting_trx_started: 2024-12-18 20:33:03
             waiting_trx_age: 00:00:23
     waiting_trx_rows_locked: 0
   waiting_trx_rows_modified: 0
                 waiting_pid: 171774
               waiting_query: drop table emails.hangfire_jobqueue
             waiting_lock_id: 7691700269:7892
           waiting_lock_mode: X
             blocking_trx_id: 7609295307
                blocking_pid: 0
              blocking_query: NULL
            blocking_lock_id: 7609295307:7892
          blocking_lock_mode: IX
        blocking_trx_started: 2024-12-17 17:33:26
            blocking_trx_age: 27:00:00
    blocking_trx_rows_locked: 2
  blocking_trx_rows_modified: 6
     sql_kill_blocking_query: KILL QUERY 0
sql_kill_blocking_connection: KILL 0
1 row in set (0.001 sec)

MariaDB [sys]> KILL QUERY 0;
ERROR 1094 (HY000): Unknown thread id: 0

UPD.2 After reboot locking is still present:

MariaDB [(none)]> select * from information_schema.INNODB_TRX \G
*************************** 1. row ***************************
                    trx_id: 7698617537
                 trx_state: LOCK WAIT
               trx_started: 2024-12-19 12:04:31
     trx_requested_lock_id: 7698617537:7931
          trx_wait_started: 2024-12-19 12:04:31
                trx_weight: 1
       trx_mysql_thread_id: 39
                 trx_query: drop database emails
       trx_operation_state:
         trx_tables_in_use: 0
         trx_tables_locked: 1
          trx_lock_structs: 1
     trx_lock_memory_bytes: 1128
           trx_rows_locked: 0
         trx_rows_modified: 0
   trx_concurrency_tickets: 0
       trx_isolation_level: REPEATABLE READ
         trx_unique_checks: 1
    trx_foreign_key_checks: 1
trx_last_foreign_key_error: NULL
          trx_is_read_only: 0
trx_autocommit_non_locking: 0
*************************** 2. row ***************************
                    trx_id: 7609295307
                 trx_state: RUNNING
               trx_started: 2024-12-19 12:01:16
     trx_requested_lock_id: NULL
          trx_wait_started: NULL
                trx_weight: 10
       trx_mysql_thread_id: 0
                 trx_query: NULL
       trx_operation_state:
         trx_tables_in_use: 0
         trx_tables_locked: 4
          trx_lock_structs: 4
     trx_lock_memory_bytes: 1128
           trx_rows_locked: 0
         trx_rows_modified: 6
   trx_concurrency_tickets: 0
       trx_isolation_level: REPEATABLE READ
         trx_unique_checks: 1
    trx_foreign_key_checks: 1
trx_last_foreign_key_error: NULL
          trx_is_read_only: 0
trx_autocommit_non_locking: 0
2 rows in set (0.000 sec)