Bug #4126 Node crashes when Data usage increased to 100%

• Submitted: 14 Jun 2004 15:03
• Modified: 20 Aug 2004 9:54
• Reporter: Lars Torstensson
• Status: Closed
• Category: MySQL Cluster: Cluster (NDB) storage engine
• Severity: S1 (Critical)
• Version: mysqlcluster-4.1.2-3.4.9-alpha-pc-linux-
• OS: Linux (Redhat AS)
• Assigned to: Pekka Nousiainen
• CPU Architecture: Any

[14 Jun 2004 15:03] Lars Torstensson

Description:
Did the following update in ndbsql:

SQL> update services set num_ip=2 where accesstype=3 and pop like 'cs%';  
Operation failed
[MySQL][ODBC driver][NDB Cluster]NDB-02060827 Out of memory in Ndb Kernel, data part - at execute without commit (in SQLExecDirect)
SQL> 

Database node 2 went down.
Cluster log:

Jun 14 14:34:29 na-gw NDB[1727]: [MgmSrvr] Node 1: Data usage increased to 100%(6250 32K pages of total 6250)
Jun 14 14:34:31 na-gw NDB[1727]: [MgmSrvr] Node 5: Node 2 Disconnected
Jun 14 14:34:31 na-gw NDB[1727]: [MgmSrvr] Lost connection to node 2
Jun 14 14:34:31 na-gw NDB[1727]: [MgmSrvr] Node 1: Arbitration check won - node group majority

Error log:
Date/Time: Monday 14 June 2004 - 14:34:30
Type of error: error
Message: Pointer too large
Fault ID: 2306
Problem data: DbtupIndex.cpp
Object of reference: DBTUP (Line: 79) 0x0000000a
ProgramName: NDB Kernel
ProcessID: 3728
TraceFile: NDB_TraceFile_5.trace
***EOM***

Node2.out
2004-06-03 13:59:49 [NDB] INFO     -- We are Node 2 with dynamic ID 3, our left neighbour is Node 1, our right is Node 1
Error handler shutting down system
Error handler shutdown completed - exiting
2004-06-03 14:00:50 [NDB] INFO     -- Ndb has terminated (pid 19307) restarting
2004-06-03 14:00:50 [NDB] INFO     -- Angel pid: 19306 ndb pid: 16898
2004-06-03 14:00:50 [NDB] INFO     -- NDB Cluster -- DB node 2

How to repeat:
Use all data memory.

[14 Jun 2004 15:06] Lars Torstensson

Trace file

Attachment: NDB_TraceFile_5.zip (application/x-zip-compressed, text), 35.36 KiB.

[20 Aug 2004 9:54] Martin Skold

Unable to reproduce, seems to be gone for 4.1.4 release