My Oracle Support Banner

Unable To Restart MySQL Cluster after [un]Planned Maintenance (Doc ID 2503283.1)

Last updated on MARCH 03, 2020

Applies to:

MySQL Cluster - Version 7.0 and later
Information in this document applies to any platform.
Whenever any investigative or analytical work is carried out on a Cluster node , care must be taken by the Admin to ensure that all file system and directory ownership / permissions are correct.

It can happen that due to maintenance activity performed , eg as root, during an outage , some files / directories may end up with incorrect permissions or ownership, which can prevent the Cluster from starting.

It may not be immediately obvious as to why this is, via the logs.

Goal

To encourage due diligence and recommended practice to ensure all maintenance / editing or copying of files etc is performed as the OS cluster user ( typically uid=mysql).

Worked Example

Scneraio

A nominal cluster environment, where an event happens which causes the cluster to be shutdown ( planed maintenance or overload condition etc ).

Events

The DBA will investigate and edit / copy some files which will end up with the incorrect ownership , eg root which are not go+r accessible to the MySQL Cluster os user uid=mysql

Effect

Unable to start the NDB Cluster

While starting the management nodes, the following errors is being received:


 

Solution

To view full details, sign in with your My Oracle Support account.

Don't have a My Oracle Support account? Click to get started!


In this Document
Goal
 Worked Example
 Scneraio
 Events
 Effect
Solution


My Oracle Support provides customers with access to over a million knowledge articles and a vibrant support community of peers and Oracle experts.