problem with heartbeat

Discussion in 'Server Operation' started by dragos, Jun 8, 2006.

  1. dragos

    dragos New Member

    I have a problem with starting heartbeat on my two load balancers. I've followed all the instructions on the HOWTO but when I'm trying to start heartbeat it doesn't work. On both loadbalancersthis error appears after I give the command "/etc/init.d/heartbeat start":

    ERROR: Bad nodename in /etc/ha.d/haresources [IPaddr2::192.168.0.105/24/eth0/192.168.0.255]
    ERROR: Configuration error, heartbeat not started.

    Help me please!!!
    I'm realy need this!
     
    Last edited: Jun 8, 2006
  2. falko

    falko Super Moderator ISPConfig Developer

    You must use the output of
    Code:
    uname -n
    as node name in /etc/ha.d/haresources.
     
  3. dragos

    dragos New Member

    Thanks!

    I've changed it the way you said and I've erased the "\" that was at the end of the line. Now there's a new error message on both machines:

    ERROR: Bad nodename in /etc/ha.d/haresources [ldirectord::ldirectord.cf]
    ERROR: Configuration error, heartbeat not started.

    Is it because of the "\" at the end of each line? I followed the exact howto instructions. Everything seems to be working except for the heartbeat. It's quite frustrating. I should mention that all the machines are virtual. I've used the VMware Workstation 5.
    What am I doing wrong?
     
  4. dragos

    dragos New Member

    P.s.

    when I put "\" after each line in the /etc/ha.d/haresources the error says:

    ERROR: Bad nodename in /etc/ha.d/haresources [IPaddr::192.168.0.105/24/eth0/192.168.0.255]

    if I erase the "\" at the end of each line the error is:

    ERROR: Bad nodename in /etc/ha.d/haresources [ldirectord::ldirectord.cf]
     
  5. falko

    falko Super Moderator ISPConfig Developer

    What's in /etc/ha.d/haresources?
     
  6. dragos

    dragos New Member

    On my first load balancer:

    server1 \
    ldirectord::ldirectord.cf \
    LVSSyncDaemonSwap::master \

    IPaddr2::192.168.0.105/24/eth0/192.168.0.255

    On the second is the same except for the first line. The first line is:

    server2 \

    server1 & server2 are the hostnames of the two load balancers.
    I've checked this with: uname -n.
     
  7. falko

    falko Super Moderator ISPConfig Developer

    The file must be exactly the same on both nodes, so it must contain server1 on both nodes, as described in the tutorial:

     
  8. dragos

    dragos New Member

  9. falko

    falko Super Moderator ISPConfig Developer

    /etc/ha.d/haresources must be

    Code:
    server1        \
            ldirectord::ldirectord.cf \
            LVSSyncDaemonSwap::master \
            IPaddr2::192.168.0.105/24/eth0/192.168.0.255
    , not

    Code:
    server1        \
            ldirectord::ldirectord.cf \
            LVSSyncDaemonSwap::master \
    
    IPaddr2::192.168.0.105/24/eth0/192.168.0.255
    The format is important!
     
  10. dragos

    dragos New Member

    you are right!
    that was the problem. I couldn't see it.
    now the heartbeat started fine, but there's a new problem with the LB1. I don't know how to best describe it. I'm going to send you an e-mail with some screen shots maybe you'll see what's wrong.
    thank you!
     
  11. falko

    falko Super Moderator ISPConfig Developer

    Please try to upgrade the MySQL client on your first load balancer.

    BTW: you can post your screenshots here, you don't need to send me emails. :)
     
  12. dragos

    dragos New Member

    I know this maybe sounds stupid but how can I best upgrade it.
    do I download a newer version of MySQL and untar it? or there's another way?
    what's the best way to do this without affecting the load-balanced cluster?
     
  13. dragos

    dragos New Member

    P.s.

    does it matter that I've installed MySQL on LB2? does it create any conflicts?
     
  14. dragos

    dragos New Member

    P.s.s.

    What I mean is that after step2 I've cloned the LB1 and modified it's host name and IP address.
     
  15. falko

    falko Super Moderator ISPConfig Developer

    Have a look at the end of http://www.howtoforge.com/loadbalanced_mysql_cluster_debian_p5
    It's described there.
     
  16. dragos

    dragos New Member

    thanks a lot!!!
    now it's working fine!!!!
     
  17. dragos

    dragos New Member

    after I follow this steps to perform setup:

    cd /tmp
    wget http://search.cpan.org/CPAN/authors/id/C/CA/CAPTTOFU/DBD-mysql-3.0002.tar.gz
    tar xvfz DBD-mysql-3.0002.tar.gz
    cd DBD-mysql-3.0002
    perl Makefile.PL
    make
    make install

    LB1 works fine but after I restart it, it's back to the usual error.
    what should I do?

    and another thing ... is there any command or program that I can use from my real machine (windows XP OS) to generate a lot of traffic on the two servers via the load balancer or the VIP 192.168.0.105?
    I saw a howto on server monitoring using munin. Can I use that to monitor my mysql servers?
     
  18. dragos

    dragos New Member

    nevermind the first problem.
    it's fine! I wasn't patient enough.
    but the second matter stands.
     
  19. falko

    falko Super Moderator ISPConfig Developer

    You can use Apache Benchmark (ab) ( http://www.cyberciti.biz/tips/howto-performance-benchmarks-a-web-server.html ), but I don't know if there's a windows version.


    Yes.
     
  20. dragos

    dragos New Member

    OK!
    thanks! I'll see what I can do. maybe I'll create another virtual machine for testing. sounds a little over the top but it's easyer to clone a machine than installing a hole new OS and a linux version of some VMware tool.
     

Share This Page