Please paste or attach your cnf file(s).
You skipped the first line of the error message, the full text is
[ERROR] InnoDB: The combined size of ib_logfiles should be bigger than
|
InnoDB: 200 kB * innodb_thread_concurrency.
|
i found the solution, but the error message is confusing....
Which part of it is confusing?
thats the whole file, the essential part
??
- * Fine Tuning
#
key_buffer = 256M
key_buffer_size = 1332M
max_allowed_packet = 500M - Bessere CPU-Auslastung mittels pool-of-threads statt one-thread-per-connection
thread_handling = pool-of-threads - Die soll MariaDB selbstständig auf die Anzahl der Cores einstellen
thread_pool_size = 8 - Nach wie viel ms soll ein neuer Thread in einem Pool generiert werden, wenn der aktive Thread nicht fertig wird
thread_pool_stall_limit = 20 - Wie viele Thread sind pro Pool sind möglich
thread_pool_max_threads = 300
thread_stack = 16K
thread_cache_size = 30
- This replaces the startup script and checks MyISAM tables if needed
- the first time they are touched
#myisam-recover = BACKUP
max_connections = 3000
max_user_connections = 500
table_cache = 1024
table_open_cache = 3072
table_definition_cache = 4096
thread_cache = 8
thread_concurrency = 8
#
- * Query Cache Configuration
#
query_cache_limit = 16M
query_cache_size = 0M
query_cache_type = 0
sort_buffer_size = 8M
#read_rnd_buffer_size = 1M
- NEU
innodb_flush_method=O_DIRECT
#innodb_log_buffer_size=6M
innodb_log_file_size=512M
#innodb_thread_concurrency = 0
innodb_flush_log_at_trx_commit = 2
#innodb_read_io_threads=64
#innodb_write_io_threads=64
query_cache_limit=256M
query_cache_size=300M
query_cache_type=0
#thread_cache = 8
#thread_concurrency = 32
tmp_table_size = 2048M
max_heap_table_size = 2048M
join_buffer_size = 10M
innodb_buffer_pool_size=25G
#innodb_additional_mem_pool_size = 20M
#innodb_log_file_size=3
innodb_log_buffer_size=8M
#innodb_file_per_table = ON
#transaction-isolation=READ-COMMITTED
#innodb_flush_log_at_trx_commit=1
#innodb_lock_wait_timeout = 300
#innodb_max_dirty_pages_pct = 90
#innodb_thread_concurrency=8
#innodb_flush_method=O_DIRECT
wait_timeout=60
interactive_timeout = 1080
#innodb_read_io_threads = 8
#innodb_write_io_threads = 8
innodb_buffer_pool_instances=8
??
this Part:
"Mär 19 23:58:46 mail.bludau-media.de mysqld[1772]: InnoDB: If you get repeated assertion failures or crashes, even
Mär 19 23:58:46 mail.bludau-media.de mysqld[1772]: InnoDB: immediately after the mysqld startup, there may be
Mär 19 23:58:46 mail.bludau-media.de mysqld[1772]: InnoDB: corruption in the InnoDB tablespace. Please refer to
Mär 19 23:58:46 mail.bludau-media.de mysqld[1772]: InnoDB: http://dev.mysql.com/doc/refman/5.6/en/forcing-innodb-recovery.html
Mär 19 23:58:46 mail.bludau-media.de mysqld[1772]: InnoDB: about forcing recovery."
it makes me fearing 
because it's repeatable
Well, users are supposed to read the whole message. If one reads only selected parts of it, it will be confusing no matter what. I think in your case the first reported error is perfectly clear:
[ERROR] InnoDB: The combined size of ib_logfiles should be bigger than
|
InnoDB: 200 kB * innodb_thread_concurrency.
|
Now, your config has many commented values so I'm not sure what was in force at the time when the problem occurred, but if it was innodb_log_file_size=3, it would certainly cause a problem.
do you patch this?
Horus Sirius, I have cleaned up the InnoDB startup and shutdown code in MySQL 5.7 and later in MariaDB Server 10.2 onwards.
In MariaDB 10.1, the parameter validation is crude. Even when InnoDB does not intentionally crash, it would not free all resources when it aborts startup, and the server could crash on shutdown. Back in 2016, I did not think that this is feasible to fix in the (already then) GA 10.1 or older versions. If you want a more pleasant experience of using InnoDB, you should upgrade to a newer MariaDB Server.
great job!
i'm CEO and SYS-ADMIN and DEVELOPER of bludau-media.de and shopste.com and also of tsecurity.de
All on the same Server
Never seen this error after changing configuration. I had panik!