[Linux-HA] heartbeat with postgresql

Andrew Beekhof andrew at beekhof.net
Sat Oct 23 01:38:04 MDT 2010


On Fri, Oct 22, 2010 at 7:23 PM, Dimitri Maziuk <dmaziuk at bmrb.wisc.edu> wrote:
> Andrew Beekhof wrote:
>
> OK, I'll post this and shut up.
>
>> Or are you really trying to claim that:
>>
>>    linuxha1 IPaddr::192.168.85.3 httpd smb
>>
>> is fundamentally less complex than
>>
>>    primitive IP ocf:heartbeat:IPaddr params ip="192.168.85.3"
>>    primitive http lsb:httpd
>>    primitive samba lsb:smb
>>    group v1-group IP http samba
>>    location prefer-ha1 v1-group inf: linuxha1
>
> Duh. Obviously, the former is one layer less complex: it doesn't have
> "group".
>
> I'm not sure if you're really trying to pull wool over my eyes or you
> have it all over yours. How about
>  """
>  cat "linuxha1 192.168.85.3 httpd smb" > /etc/ha.d/haresources
>  /etc/init.d/heartbeat start
> on both nodes
>  """
> vs
>  """

cat "primitive IP ocf:heartbeat:IPaddr params ip=192.168.85.3
    primitive http lsb:httpd
    primitive samba lsb:smb
    group v1-group IP http samba
    location prefer-ha1 v1-group inf: linuxha1" | crm configure

on a single node

>  where exactly do all these lines go? -- cib.xml? Where are the tags,
> then?

I think you're confusing "different" with "more complex".
Or is there some innate knowledge we all carry that tells us where
haresources lives and how to configure it?

> Some other file you need to run through something before you stat
> the cluster? Are they commands you type in some shell after you start
> the cluster? WFT is 'ocf:' and why is it different from 'lsb:'. Or
> 'inf:'? WTF does it take 1 minute to type up what previously took 15
> seconds (@20WPM)? WTF is 'prefer-ha1'? -- why is not just 'prefer', is
> there a 'prefer-ha2'? Why is '-ha1' not referenced anywhere else in the
> config? Exactly which part of all this says 'run ifup eth0:0, then start
> httpd, then samba on linuxha1' in one simple easy to read sentence?
>  """

haresources doesn't say that either.
Sure it looks like it does, and it will look like it works, right up
to the point at which it doesn't.
Just like clusters without stonith look like they work right up to the
point where they corrupt all your data.

The simplest example... v1 only checks the status of the first element
of the group.
So if somehow the IP already happens to be up, it will make no effort
to start any of the other resources.

If you've not hit this one personally, then congratulations.
Plenty of people did, me included.

>
> It's not about syntax. It's not about advanced features. It's about a
> new user setting up a simple stupid active/passive failover pair.

The fact that you can describe something in under a dozen words don't
make it any less complex.

   "I just want to go to the moon"



More information about the Linux-HA mailing list