High reliability web server - cluster, redundancy, etc Post: 302302261

Sponsored Content

Special Forums UNIX and Linux Applications High Performance Computing High reliability web server - cluster, redundancy, etc Post 302302261 by kipperjp on Monday 30th of March 2009 01:38:16 PM

03-30-2009

Registered User

Keep in mind that clustering does not mean high availability. CLustering is sharing resources with multiple machines. What you need to look into is syncing and failover mechanisms so that 2 web servers at different locations are mirrored copies so that if one goes down, dns is routed to the other server.

I am kind of down this route myself and learning as I go. Be nice to utilize an HA consultant as trial and error can be costly.

kipperjp

View Public Profile for kipperjp

Find all posts by kipperjp

3 More Discussions You Might Find Interesting

1. Solaris

High Availability zone on Sun Cluster

HI Experts, Could some one help me in configuring high availability zone on Sun Cluster Reg: Sudhan

2. Red Hat

Red Hat High Availability (HA) Cluster

How can we implement a service in HA, which in not available in HA. like sldap or customize application. Requirement Details. NODE1 service slapd is running.(Require) NODE2 service slapd is running.(Require) on both the node replication is happening. Now here requirement is need...

3. Red Hat

Web server cluster at some point ?

What's the best way clusters for Storage at some point? (The way that data is the same in all parts) To set up a Web server cluster is the logical way?! Cluster database and Storage and then by keepalived + HA cluster will be communicated? Or, there a better solution? (For about 4 points) Thank...

LEARN ABOUT DEBIAN

kumoctl

KUMOCTL(1)						      General Commands Manual							KUMOCTL(1)

NAME

       kumoctl - show kumofs status

SYNOPSIS

       kumoctl address[:port=19750] command [options]

DESCRIPTION

       Show status of kumo-manager and send control commands to kumo-manager.

COMMANDS

       status show status of the cluster

       attach attach all new servers and start replace

       attach-noreplace
	      attach all new servers

       detach detach all fault servers and start replace

       detach-noreplace
	      detach all fault servers

       replace
	      start replace without attach/detach

       full-replace
	      start full-replace (repair consistency)

       backup [suffix=20090304]
	      create backup with specified suffix

       enable-auto-replace
	      enable auto replace

       disable-auto-replace
	      disable auto replace

STATUS

       hash space timestamp
	      The  time  that the list of attached kumo-servers is updated. It is updated when new kumo-server is added or existing kumo-server is
	      down.

       attached node
	      The list of attached kumo-servers. (active) is normal node and (fault) is fault node or recoverd but not re-attached node.

       not attached node
	      The list of recognized but not-attached nodes.

AUTO REPLACING

       By default, you have to attach new kumo-servers manually using kumoctl command. If the auto replacing  is  enabled,  new  kumo-servers  are
       attached automatically.	This is experimental feature.

EXAMPLE

       $ kumoctl mgr1 status

       $ kumoctl mgr1 attach

SEE ALSO

       kumo-manager(1).

kumoctl 							     July 2009								KUMOCTL(1)