MySQL Forums
Forum List  »  NDB clusters

Re: Wait LCP to ensure durability
Posted by: Mikael Ronström
Date: November 27, 2017 08:44AM

The IO threads are using count=1, this is a bit of a misnomer.
There is a lot of IO threads, one for each open file at least
and in addition by default 2 threads for tablespace files.
In the configuration it is mandatory to set count, so it must
be set to 1, for IO threads we actually ignore this number.

These threads are most likely the threads with tid's a bit
higher (in the 8000 and 9000 ranges).

You can get even better understanding which are the IO threads
by ensuring that top prints the last CPU the thread executed on.
The IO threads should all have executed last on CPU 11.

Options: ReplyQuote


Subject
Views
Written By
Posted
632
September 28, 2017 03:46AM
264
September 28, 2017 03:50AM
337
September 28, 2017 03:17PM
397
October 02, 2017 01:56AM
280
October 02, 2017 06:39AM
249
October 04, 2017 03:49AM
247
November 22, 2017 01:07AM
217
November 22, 2017 01:35AM
198
November 22, 2017 02:23AM
250
November 22, 2017 03:30AM
271
November 25, 2017 07:33AM
217
November 27, 2017 04:28AM
Re: Wait LCP to ensure durability
193
November 27, 2017 08:44AM
201
November 27, 2017 04:16PM


Sorry, you can't reply to this topic. It has been closed.

Content reproduced on this site is the property of the respective copyright holders. It is not reviewed in advance by Oracle and does not necessarily represent the opinion of Oracle or any other party.