Bug #25237 | SELECT causes data node crash | ||
---|---|---|---|
Submitted: | 21 Dec 2006 16:03 | Modified: | 10 Oct 2007 7:59 |
Reporter: | Victoria Reznichenko | Email Updates: | |
Status: | Can't repeat | Impact on me: | |
Category: | MySQL Cluster: Cluster (NDB) storage engine | Severity: | S1 (Critical) |
Version: | 5.0.34bk | OS: | Linux (SuSe) |
Assigned to: | Martin Skold | CPU Architecture: | Any |
[21 Dec 2006 16:03]
Victoria Reznichenko
[21 Dec 2006 16:04]
MySQL Verification Team
logs and trace files
Attachment: ndb_error_report_20061221185726.tar.bz2 (application/bzip2, text), 714 bytes.
[27 Dec 2006 12:41]
MySQL Verification Team
logs and trace
Attachment: bug25237_logs_trace.zip (application/zip, text), 45.61 KiB.
[16 May 2007 9:45]
Martin Skold
The problem seems to be resource shortage that is not handled properly. Increasing TransactionBufferMemory (1M by default) should solve the problem. The bug fix will be that the transaction is aborted with an error message telling the user that this parameter needs to be increased.