NDBD Fails To Restart

[Huseyin Varol]

Hi,
I searched the forum and the list yet i did not find anyone talking about this (actually someone has the same error but my situation is not the same). I was testing StopOnError condition. I use StopOnError=0 in my config.ini NDBD sections, so I expect my NDBD to restart after a connection failure. Yet in my tests it fails. Here are the details:

I run first MGM and NDBD on one computer. Then my test NDBD on a second computer. I do not run second MGM and any API node. And this error does not occur if i delete the second MGM from config.ini and use ndb-connectstrings accordingly when calling processes. But it should work on these conditions right? However, it does not, and when i disconnect the second computer and then reconnect it, the ndbd process fails.

The commands i use sequentially:
On computer 1 (10.11.12.158):
/usr/sbin/ndb_mgmd (in the same directory with recent config.ini)
/usr/sbin/ndbd --initial --ndb-connectstring=nodeid=3,10.11.12.158,10.11.12.209
On computer 2 (10.11.12.15):
/usr/sbin/ndbd --initial --ndb-connectstring=nodeid=4,10.11.12.158,10.11.12.209

***config.ini***
[NDBD DEFAULT]
NoOfReplicas=2
[mysqld default]
[ndb_mgmd default]
[tcp default]

[NDB_MGMD]
Id=1
Hostname=10.11.12.158

[NDB_MGMD]
Id=2
Hostname=10.11.12.209

[NDBD]
Id=3
Hostname=10.11.12.158
datadir=/home/huseyin/ndbdata/
StopOnError=0

[NDBD]
Id=4
HostName=10.11.12.15
StopOnError=0

[MYSQLD]
Id=5
[MYSQLD]
Id=6

***The error log (ndb_4_error.log)***
Current byte-offset of file-pointer is: 468


Date/Time: Monday 22 August 2005 - 14:32:10
Type of error: error
Message: Arbitrator shutdown
Fault ID: 2305
Problem data: Arbitrator decided to shutdown this node
Object of reference: QMGR (Line: 3796) 0x00000008
ProgramName: /usr/sbin/ndbd
ProcessID: 2227
TraceFile: ./ndb_4_trace.log.1
Version 4.1.13
***EOM***

***The out log(ndb_4_out.log)***
2005-08-22 14:31:47 [NDB] INFO -- Angel pid: 2226 ndb pid: 2227
2005-08-22 14:31:47 [NDB] INFO -- NDB Cluster -- DB node 4
2005-08-22 14:31:47 [NDB] INFO -- Version 4.1.13 --
2005-08-22 14:31:47 [NDB] INFO -- Configuration fetched at 10.11.12.158 port 1186
Error handler restarting system
Error handler shutdown completed - exiting
2005-08-22 14:32:48 [NDB] INFO -- Ndb has terminated (pid 2227) restarting
2005-08-22 14:32:51 [NDB] INFO -- Angel pid: 2226 ndb pid: 2680
2005-08-22 14:32:51 [NDB] INFO -- NDB Cluster -- DB node 4
2005-08-22 14:32:51 [NDB] INFO -- Version 4.1.13 --
2005-08-22 14:32:51 [NDB] INFO -- Configuration fetched at 10.11.12.158 port 1186
NR: setLcpActiveStatusEnd - m_participatingLQH

That's all.
Thanks for any help
Huseyin