2

I have the cluster built by one MGM Node and two data/sql nodes

When importing data to an NDB node the importation crashes returning the following:

ndb_2# mysql -u root -p db < data.sql
Enter password:
ERROR 1297 (HY000) at line 214: Got temporary error 4010 'Node failure caused abort of transaction' from NDBCLUSTER

Monitoring the other NDB node, I could see the data is being well inserted, but at 'n' records, the first node return the ERROR.

The ndb_2_out.log & ndb_3_out.log write:

WARNING  -- ACK wo/ gcp record (gci: 60138/0) ref: 0fa20004 from: 0fa20004

The ndb_2_error.log & ndb_3_error.log don't write anything.

Before the crash, both nodes have the same records.

I google it but I can't find anything related.

2

By reading some forums, the variables that avoids this issue are:

MaxBufferedEpochs
TimeBetweenEpochsTimeout

This must be configured in MGMT Node

/var/lib/mysql-cluster/config.ini

As we can see in MySQL Documentation, the default for MaxBufferedEpochs is 100, the default for TimeBetweenEpochsTimeout is 4000, change this for the needed value.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.