HowtoForge Forums | HowtoForge - Linux Howtos and Tutorials - problem with load-balanced mysql cluster

HowtoForge Forums | HowtoForge - Linux Howtos and Tutorials (http://www.howtoforge.com/forums/index.php)

- HOWTO-Related Questions (http://www.howtoforge.com/forums/forumdisplay.php?f=2)

- - problem with load-balanced mysql cluster (http://www.howtoforge.com/forums/showthread.php?t=7297)

problem with load-balanced mysql cluster

I'm having some problem with configuren load-balanced mysql-cluster setup.

I've followed the setps in How to set up A load-balanced Mysql Cluster.

The first problem is with hearbeat. A get te following error:

Code:

heartbeat: 2006/10/06_16:00:01 ERROR: mcast [eth0] bad addr [255.0.0.1]

heartbeat: 2006/10/06_16:00:01 ERROR: Illegal directive [respawm] in /etc/ha.d/ha.cf

heartbeat: 2006/10/06_16:00:01 ERROR: Heartbeat not started: configuration error.

heartbeat: 2006/10/06_16:00:01 ERROR: Configuration error, heartbeat not started.

when I comment these lines out in the ha.cf heartbeat will start and it looks like the load-balancers are running and all tests in the how-to are showing the right response.
But when I want te connect from a other mahine with a mysql-client, I can't connect. After a portscan on the virtual ip of the loadbalancer I see that te mysql port is filtered (I'm not using a firewall).

I've really no idea what is wrong? And if the first problem affects the second problem that I can't connect.

This is my ha.cf file:

Code:



logfacility     local0

bcast           eth0

mcast eth0 255.0.0.1 694 1 0

auto_failback   off

node            loadb1

node            loadb2

respawm hacluster /usr/lib/heartbeat/ipfail

apiauth ipfail gid=haclient uid=hacluster

Some help will be very nice! :D

Quote:

Originally Posted by MvdB

Code:



logfacility     local0

bcast           eth0

mcast eth0 255.0.0.1 694 1 0

auto_failback   off

node            loadb1

node            loadb2

respawm hacluster /usr/lib/heartbeat/ipfail

apiauth ipfail gid=haclient uid=hacluster

You have a typo there. It must be

Code:

respawn hacluster /usr/lib/heartbeat/ipfail

Aha Tnx! :P

But has the error:

Code:

heartbeat: 2006/10/06_16:00:01 ERROR: mcast [eth0] bad addr [255.0.0.1]

Anything to do with this typo?

Quote:

Originally Posted by MvdB

Aha Tnx! :P

But has the error:

Code:

heartbeat: 2006/10/06_16:00:01 ERROR: mcast [eth0] bad addr [255.0.0.1]

Anything to do with this typo?

You have another error in your ha.cf. It must be

Code:

mcast eth0 225.0.0.1 694 1 0

instead of

Code:

mcast eth0 255.0.0.1 694 1 0

Tnx man!

It works :D Great! :D

It were the 2 typo's and a config error with iproute on the sql servers.

:p