Johan:
WOW, this is just amazing, seems a little complicated to configure, but
nothing in Linux is for the faint of heart :), still I wish for the
MySQL ODBC to have a fallback feature in a not so distant future. (is it
available in the new ODBC alfa Mikael?), Maybe in the MySQL Cluster
5.0?? Anyway... great job guys!
Alejandro Ramírez
-----Mensaje original-----
De: Johan Andersson [mailto:johan@stripped]
Enviado el: Jueves, 01 de Julio de 2004 12:52 p.m.
Para: 'Alejandro J. Ramírez'; 'Mikael Ronström'
CC: cluster@stripped
Asunto: RE: ODBC and MySQL Cluster?
Hi,
perhaps you can look at http://www.linuxvirtualserver.org/ to get som
inspiration on how to solve your problem.
Johan Andersson
> -----Original Message-----
> From: Alejandro J. Ramírez [mailto:aramirez@stripped]
> Sent: den 1 juli 2004 18:26
> To: 'Mikael Ronström'
> Cc: cluster@stripped
> Subject: RE: ODBC and MySQL Cluster?
>
> Great Answer, thanks a lot Mikael!
>
>
> Alejandro Ramírez
>
> -----Mensaje original-----
> De: Mikael Ronström [mailto:mikael@stripped] Enviado el:
> Jueves, 01 de Julio de 2004 09:11 a.m.
> Para: Alejandro J. Ramírez
> CC: cluster@stripped
> Asunto: Re: ODBC and MySQL Cluster?
>
> Hi Alejandro,
>
> 2004-07-01 kl. 17.48 skrev Alejandro J. Ramírez:
>
> > Hi Mikael,
> >
> > So if the API goes down it is not transparent for the
> application
> > that needs to run on the Cluster? For example people were talking
> > about having a webserver on each node... if one webserver
> goes down,
> > you would have to have some rule (metric?) that tells
> people trying to
> > get to that webserver, that they should try other ip address (Other
> > webserver on the cluster) for reaching that same service?
> This might
> > be a little out of the scope of the mailing list but I am trying to
> > figure out a useful way of using the cluster.
> >
>
> If a web server goes down I presume there are tools, HW and
> so forth available to balance over the load on other web
> servers so I think integrating MySQL servers with web server
> should work out fine even without failover between MySQL
> client and MySQL server. Not my expertise area though.
>
> Using Connector/J (MySQL JDBC driver) fail-over to another
> MySQL server is already supported. It can be done fairly
> easily using Perl DBI. MySQL .Net Connector will get the
> feature in a not so distant future and at some point in time
> it will be integrated into libmysql such that the remaining
> MySQL connectors will also support fail-over to another MySQL
> server at failure of the MySQL server connection.
>
> Rgrds Mikael
>
>
> > Thanks!
> >
> > Alejandro Ramírez
> >
> >
> >
> >
> >
> > -----Mensaje original-----
> > De: Mikael Ronström [mailto:mikael@stripped] Enviado el:
> Jueves, 01
> > de Julio de 2004 08:32 a.m.
> > Para: Alejandro J. Ramírez
> > CC: cluster@stripped
> > Asunto: Re: ODBC and MySQL Cluster?
> >
> > Hi Alejandro,
> >
> > 2004-07-01 kl. 17.09 skrev Alejandro J. Ramírez:
> >
> >>
> >>
> >> Hi All,
> >>
> >> I´ve managed to successfully install the cluster in
> Redhat
> >> 9.0 (BTW Bison 1.75 does not do the trick in RH9
use 1.85,
> the Fedora
> >> RPM will do..) however I am a little confused as how the
> API works in
> >> the cluster. How would I go about using ODBC with the cluster? How
> >> should my app know if an API node is down to use another
> one?? As you
> >> may see I am very confused
this is my first cluster
> (newbie..) and I
> >> am very excited to get it to do something useful.
> >>
> >
> > If API node is down it means that this particular MySQL server is
> down.
> > If MySQL
> > server is down your connections to that server should be
> down. So it
> > should be discovered by some kind of error code from ODBC
> saying that
> > connection is broken.
> > Unfortunately not an expert in ODBC on that level so don't know
> exactly
> > where you
> > get error messages like that.
> >
> > Rgrds Mikael
> >
> >>
> >> One more thing for everyone else trying to get the
> cluster
> >> to work.. there is a thread mostly Tomas and Matteo there
that
> >> explains how the /etc/hosts files should be configured, I found it
> >> rather confusing at first but then realized what Matteo
> was saying.
> >> In
> short:
> >>
> >> MGMT Node Cluster1
> >>
> >> 127.0.0.1 localhost
> >> 10.34.0.1 Cluster1
> >> 10.34.0.2 Cluster2
> >>
> >> DB Node Cluster2
> >> 127.0.0.1 localhost
> >> 10.34.0.1 Cluster1
> >> 10.34.0.2 Cluster2
> >>
> >> Simple huh? But it takes a while to guess it
> >>
> >> Thanks,
> >>
> >>
> >> Alejandro Ramírez
> >>
> > Mikael Ronström, Senior Software Architect MySQL AB, www.mysql.com
> >
> > Clustering:
> > http://www.infoworld.com/article/04/04/14/HNmysqlcluster_1.html
> >
> > http://www.eweek.com/article2/0,1759,1567546,00.asp
> >
> >
> >
> >
> >
> Mikael Ronström, Senior Software Architect MySQL AB, www.mysql.com
>
> Clustering:
> http://www.infoworld.com/article/04/04/14/HNmysqlcluster_1.html
>
> http://www.eweek.com/article2/0,1759,1567546,00.asp
>
>
>
> --
> MySQL Cluster Mailing List
> For list archives: http://lists.mysql.com/cluster To unsubscribe:
> http://lists.mysql.com/cluster?unsub=1
>
>
>
> --
> MySQL Cluster Mailing List
> For list archives: http://lists.mysql.com/cluster
> To unsubscribe:
> http://lists.mysql.com/cluster?unsub=1
>
>
>
