For test purpose, I'm running MySQL 4.1.10a with cluster configuration
on Linux(ndbd node: 2.6.8-4smp) for 2 weeks.
I have troubles every day or night. abruptly 1 or 2 nodes
of 2 ndbd nodes have downed on no query workload.
Error log is
====================================================
Date/Time: Monday 4 April 2005 - 09:59:07
Type of error: error
Message: Job buffer congestion
Fault ID: 2334
Problem data: Job Buffer Full
Object of reference: APZJobBuffer.C
ProgramName: ndbd
ProcessID: 16818
TraceFile: /var/lib/mysql-cluster/ndb_3_trace.log.2
***EOM***
====================================================
Some times, restarting successful. Some times restarting, but after every
5 minites, that node is down agein and restarting. So, 5 times or so it
disconnected for ever. Ndb daemon is alive in silence.
Do you know this type of failure and that cause?
What is job Buffer? Which config parameter is related?
Is this memory problem, LAN problem, table definition problem,
or program bug?
How can I get rid of these failure?
My setup is normal 2 nodes setup.
====================================================
[NDBD DEFAULT]
NoOfReplicas=2
[MYSQLD DEFAULT]
[NDB_MGMD DEFAULT]
[TCP DEFAULT]
[NDB_MGMD]
HostName=172.16.2.12
[NDB_MGMD]
HostName=172.16.2.11
[NDBD]
HostName=172.16.2.31
DataDir=/var/lib/mysql-cluster
StopOnError=false
[NDBD]
HostName=172.16.2.32
DataDir=/var/lib/mysql-cluster
StopOnError=false
[MYSQLD]
[MYSQLD]
====================================================
-----------
Nobo Furuya
