Re: Wait LCP to ensure durability
To change cpu binding for NDB can only be done at restart and
I presume you don't want to restart the nodes during the restore.
The other way of changing CPU binding is to use top -H to discover
the threads that are using CPU 12 and next use taskset or numactl
to specifically bind those to other CPUs.
If you do this, the CPU locking will remain until the node is
restarted again, so you might want to consider binding them back
to the original CPU again when the restore process is done.
Subject
Views
Written By
Posted
2333
September 28, 2017 03:46AM
967
September 28, 2017 03:50AM
1163
September 28, 2017 03:17PM
1781
October 02, 2017 01:56AM
1073
October 02, 2017 06:39AM
1078
October 04, 2017 03:49AM
1214
November 22, 2017 01:07AM
1051
November 22, 2017 01:35AM
1028
November 22, 2017 02:23AM
1017
November 22, 2017 03:30AM
967
November 23, 2017 05:06AM
1012
November 25, 2017 02:33AM
1073
November 25, 2017 04:08AM
1077
November 25, 2017 04:16AM
987
November 25, 2017 07:33AM
907
November 25, 2017 11:44AM
945
November 25, 2017 03:37PM
877
November 26, 2017 01:00AM
1156
November 26, 2017 08:06AM
Re: Wait LCP to ensure durability
940
November 27, 2017 04:28AM
996
November 27, 2017 04:58AM
922
November 27, 2017 08:44AM
984
November 27, 2017 09:16AM
947
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.