[Linux-cluster] Re: RFC: updating cluster.conf

[denis]

David Teigland wrote:
> Hi,
> 
> We're looking into how cluster.conf updates should be done in future
> versions and we'd like some feedback about how you currently do this, and
> what you'd like to see.
> 
> 1. How often do you update cluster.conf?  ("Never" would be valuable
>    feedback.)

After stabilizing the config, ~never, but probably will be something 
like 1 per 6months. Prior to setting the system in production, very 
frequent updates (several per day).

> 2. What changes do you make?  e.g. add nodes, change fencing settings,
>    add or change rgmanager settings.

During test, change fencing settings, modify services / hosts.

> 3. How do you currently update cluster.conf?  Cluster online or offline?
>    Manually scp to all nodes?  ccs_tool?  conga?  What do you like and not
>    like about the method you use now?

Cluster online:
  1. Edit cluster.conf.new in Puppet working copy
  2. Commit to Puppet SVN
  3. "puppetd -t" on all nodes or verify that cluster.conf.new is 
updated on all nodes if Puppet is running automatically.
  4. "ccs_tool update /etc/cluster/cluster.conf.new"

Like about it :
  * Centralized storage of all configuration in Puppet
  * Full control of update process
  * Revision control of configuration makes it easier to debug errors 
due to changes

Areas that could be improved :
  * Very manual process

> 4. How would you like to do updates to cluster.conf in the future?
>    Conga (graphical management interface)?  Command line program that
>    updates /etc/cluster/cluster.conf on all cluster nodes?
>    Manually scp to all nodes?  Other?

Would be great if I could have some sort of automated loading of 
configuration changes when all nodes have the correct version of the 
config.

> 5. Would you like to use an LDAP server?  All cluster nodes would read
>    cluster.conf info from the server; updates would just be made on
>    the server.

Maybe. Not sure if the added complexity would be worthwhile in this 
small setup. For larger clusters I understand the benefits.

Should at least be an optional way of doing things imho.

Regards
--
DenisB