Re: fail-over using vrrp and MISC_CHECK

"Yu Safin" <[email protected]> Tue, 31 Oct 2006 13:22:12 -0500
Newsgroups gmane.linux.keepalived.announce
Message-ID <[email protected]>
On 10/27/06, Yu Safin <[email protected]> wrote:
> I have spent quite a bit of time in google trying to find samples on
> how to code my own MISC_CHECK.  My second quest is to find where and
> how I can code the call to my MISC_CHECK inside my KEEPALIVED.CONF
> files.  The samples for TCP_CHEKC and HTTP_GET I have found apply to
> real-servers.  I am not using such parameter but I suspect I will have
> to.
>
> I have set up KeepAlived to provide me with a fail-over capability
> between two servers running my own application.  The data in the
> application gets syncronized between both servers while some other
> data is kept via NFS mounted on both servers (exported from a 3rd
> server).
>
> Both servers have only one Ethernet Interface (ETH0) under SLES 10.0.
> What I would like to do is write my own perl script to check on
> various conditions to determine if the server is functioning or not.
>  From reading the user guide, I can send a return code of -1 to
> indicate the TEST FAILED so the fail-over can take place.
>
>
> This is a sample for the MASTER server /etc/keepalived/keepalived.conf file.
>
> vrrp_instance VI_1 {
>  state MASTER
>  interface eth0
>  virtual_router_id 1
>  priority 100
>  authentication {
>  auth_type PASS
>  auth_pass concha
>  }
>  virtual_ipaddress {
>  192.168.10.101/24 brd 192.168.10.255 dev eth0
>  }
> }
>
> This is a sample for the BACKUP server /etc/keepalived/keepalived.conf file.
>
> vrrp_instance VI_1 {
>  state BACKUP
>  interface eth0
>  virtual_router_id 1
>  priority 50
>  authentication {
>  auth_type PASS
>  auth_pass concha
>  }
>  virtual_ipaddress {
>  192.168.10.101/24 brd 192.168.10.255 dev eth0
>  }
> }
>
> Notice that the VIP is 192.168.10.101.  MASTER is at 192.168.10.183
> and BACKUP is at 192.168.10.200.
>
I ended up setting up a daemon to check on my applications and when
problems are detected on the MASTER, then the eth0 interface is
brought down.
When the daemon detects that the situation was corrected (or after a
reboot), then the interface is brought up again and MASTER regain
control.

-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642