Last updated on OCTOBER 31, 2016
Applies to:MySQL Cluster - Version 7.1 and later
Information in this document applies to any platform.
- Initial condition: a small amount of traffic running (i.e. less that 20x the normal traffic requests)
- Reboot half of the data nodes in order to stop half of the Cluster
During the reboot, the process handling the traffic (and performing cluster request) have some difficulties to retrieve data.
In the core stack file retrieved, it appears that most of the thread are blocked in pthread_cond_timedwait in ndb API.
How to explain this performance degradation?
Sign In with your My Oracle Support account
Don't have a My Oracle Support account? Click to get started
Million Knowledge Articles and hundreds of Community platforms