Hi Percona Team,
We run a Ubuntu 14.04 based 500GB+ sized mysql5.5 database with about 500+ tables in our production environment. We had a year-end data migration based on some table/schema changes for our database on 30th December 2017. To have minimal downtime, the migration was done in another server and a full dump using innobackupex version 2.3.9 was generated and brought back to the main production server and the database was restored.
Since then, in the production server the CPU usage is always around 90-100% and the load average is quite high., that we have to restart mysqld every 2 hours or so. We are hosted on Amazon ec2 r4.2xlarge instance which has 61MB RAM and 8 CPUs (amazon AWS notion equivalent to 27 core).
As per some advise, ran “show innodb engine status” but not sure if there anything amiss there.
Appreciate your inputs and how we can progress further in this scenario.