error 293 Inconsistent trigger state in TC block

[Pete Ditmars]

Hi,

I'm trying to run some tests with our mysql NDB cluster and we initially got some of the following errors:

Got temporary error 221 'Too many concurrently fired triggers (increase MaxNoOfFiredTriggers)'

And so, following the guidance, we increased the MaxNoOfFiredTriggers, and following that we got

Got temporary error 233 'Out of operation records in transaction coordinator (increase MaxNoOfConcurrentOperations)'

So we've increased that parameter as well. Currently with these settings

[ndbd]
MaxNoOfConcurrentOperations=6553600
MaxNoOfFiredTriggers=4000000

we're getting neither of the above errors, but now we're getting this error, and I haven't found any guidance in any forums or documentation on what is causing this or what to try:

Got error 293 'Inconsistent trigger state in TC block' from NDBCLUSTER

Does anyone have any insight into what this error means and what I can do about it?

Thanks very much,

Pete

p.s.

We're running this version of MySQL

Server version: 5.7.16-ndb-7.5.4-cluster-commercial-advanced MySQL Cluster Server - Advanced Edition (Commercial)

We've seen this on inserts and on deletes of a number of records. The slow query log is showing us this for the delete queries that are encountering the 'Inconsistent trigger state in TC block' errors.

# Time: 2018-03-14T21:48:13.716121Z
# User@Host: user[user] @ localhost [] Id: 33277
# Query_time: 20.595956 Lock_time: 0.000161 Rows_sent: 0 Rows_examined: 100512
SET timestamp=1521064093;
delete s, ao from S_TABLE s JOIN A_TABLE ao ON s.OBJ = ao.ID where time <= 1521042709062;



Edited 2 time(s). Last edit at 03/14/2018 04:05PM by Pete Ditmars.