From: Johan Andersson
Date: December 1 2010 6:56am
Subject: Re: Question concerning ndbd --initial
List-Archive: http://lists.mysql.com/cluster/7906
Message-Id: <5AE62DC6-0E10-41E0-BEFC-D6E1A6ED4C77@severalnines.com>
MIME-Version: 1.0 (Apple Message framework v1082)
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable

Hi,

Here is the procedure in the ref manual for  what you want to do:
=
http://dev.mysql.com/doc/refman/5.0/en/mysql-cluster-rolling-restart.html

Some config changes (redo log, datadir) requires that you restart the =
data node with --initial.

If you run the --initial on one data node at the time you are safe, it =
will only clears its local filesystem to recreate redo log files e.g

if you run --initial on all data nodes at the same time, you will clear =
the data.


BR
johan


1 dec 2010 kl. 07.26 skrev Peter S=F8rensen:

> Hi,
>=20
> I have found the following procedure to handle a change in the cluster =
configuration file:
> -------------------------------------------------------
> 5 How To Restart The Cluster
> Now let's asume you want to restart the MySQL cluster, for example =
because you have changed /etc/mysql/ndb_mgmd.cnf on mxdb-mgr.sdu.dk or =
for some other reason. To do this, you use the ndb_mgm cluster =
management client on mxdb-mgr.sdu.dk:
> mxdb-mgr.sdu.dk:
>=20
> # ndb_mgm
> On the ndb_mgm console, you type shutdown;=20
> You will then see something like this:
> ndb_mgm> shutdown;
> Node 3: Cluster shutdown initiated
> Node 2: Node shutdown completed.
> 2 NDB Cluster node(s) have shutdown.
> NDB Cluster management server shutdown.
> ndb_mgm>
>=20
> This means that the cluster data nodes mxdb1.sdu.dk and mxdb2.sdu.dk =
and also the cluster management server have shut down.
> Type quit;
> to leave the ndb_mgm console.=20
>=20
> To start the cluster management server, do this on mxdb-mgr.sdu.dk:
> mxdb-mgr.sdu.dk:
> # rm /usr/mysql-cluster/ndb_1_config.bin.1 (nessacary to remove the =
cached config file)
> # service mysql-ndb-mgm start
> and on mxdb1.sdu.dk and mxdb2.sdu.dk you can run
> mxdb1.sdu.dk / mxdb2.sdu.dk:=20
> ndbd
> or, if you have changed /etc/mysql/ndb_mgmd.cnf on mxdb-mgr.sdu.dk:
> ndbd --initial
> Afterwards, you can check on mxdb-mgr.sdu.dk if the cluster has =
restarted:
> mxdb-mgr.sdu.dk:
> # ndb_mgm
> On the ndb_mgm console, type=20
> show;
> to see the current status of the cluster. It might take a few seconds =
after a restart until all nodes are reported as connected.=20
> Type quit;
> to leave the ndb_mgm console. =20
> =
--------------------------------------------------------------------------=
---------
>=20
> I am uncertain on the ndbd --initial . Won't that reset all my =
clustered data ??
>=20
> Regards
>=20
> Peter S=F8rensen/Univ. of Southern Denmark/email: maspsr@stripped
>=20
> --=20
> MySQL Cluster Mailing List
> For list archives: http://lists.mysql.com/cluster
> To unsubscribe:    =
http://lists.mysql.com/cluster?unsub=3Djohan@stripped
>=20
>=20