Hi,
The trace files indicates that heartbeat failures might have occured.
Could you check the cluster log?
If the importing creates so much load so that heartbeats gets missed,
this can be fixed by modifying the "HeartbeatIntervalDbDb" and/or
"HeartbeatIntervalDbApi".
FYI: we're working on improving the heartbeat handling so it won't be load dependant...
/Jonas
Nathan J. Mehl wrote:
> Folks--
>
> Well, this is not entirely promising. I'm managing to get a repeated
> abort of ndbd while trying to import a database from a dumpfile.
>
> The error and trace files can be found here:
>
> http://blank.org/memory/work/ndbd-error/
>
> The details:
>
> mysql-max-4.1.11-pc-linux-gnu-i686 from the dev.mysql.com tarball
> CentOS 4 (RedHat Enterprise 4.0 clone-from-SRPM; see www.centos.org)
> Linux lokar 2.6.9-5.0.3.ELcustom #1 Tue Apr 19 21:51:03 EDT 2005 i686 athlon i386
> GNU/Linux
>
> my.cnf:
>
> [mysqld]
> ndbcluster
> ndb-connectstring=spaceghost
> [mysql_cluster]
> ndb-connectstring=spaceghost
>
> config.ini:
>
> [NDBD DEFAULT]
> NoOfReplicas=2
> [MYSQLD DEFAULT]
> [NDB_MGMD DEFAULT]
> [TCP DEFAULT]
> # Managment Server
> [NDB_MGMD]
> HostName=spaceghost # the IP of THIS
> SERVER
> # Storage Engines
> [NDBD]
> HostName=lokar # the IP of the FIRST SERVER
> DataDir= /project/mysql-cluster
> DataMemory=512M
> IndexMemory=128M
> MaxNoOfConcurrentOperations=250000
> [NDBD]
> HostName=brak # the IP of the SECOND SERVER
> DataDir=/project/mysql-cluster
> DataMemory=512M
> IndexMemory=128M
> MaxNoOfConcurrentOperations=250000
> [MYSQLD]
> [MYSQLD]
>
>
> Any suggestions anyone could offer here would be well appreciated.
>
> -n
>
> ------------------------------------------------------------<memory@stripped>
> "The beauty you like is precisely that which escapes you." (--Issey Miyake)
> <http://blank.org/memory/>----------------------------------------------------
>
--
Jonas Oreland, Software Engineer
MySQL AB, www.mysql.com
