RH/CentOS network configurations

[jarod-ajLrJawYSntWk0Htik3J/w at public.gmane.org (Jarod Wilson)]

On Sep 9, 2009, at 12:07 PM, theBlueSage wrote:

> On Tue, 2009-09-08 at 16:43 -0400, Jarod Wilson wrote:
>> On Sep 8, 2009, at 2:17 PM, theBlueSage wrote:
>>
>>> okay, scratch that, I figured out that pre-up sysctl -p simply
>>> performed
>>> a /sbin/sysctl reread before bringing up the interface, si I have  
>>> that
>>> set and now unnecessary.
>>> Having said that, if anyone has got good info/tutorials on how to  
>>> use
>>> heartbeat/ldirectord in a Centos Environment I love to see them as
>>> this
>>> is proving frustratingly undocumented. Centos, platypus-like as  
>>> ever,
>>> may look like Red Hat, but Red Hat it is not, especially when it  
>>> comes
>>> to LB/cluster work .. :(
>>
>> Um... Unless you're using a bunch of CentOS Plus packages and/or 3rd-
>> party repos, there shouldn't be any significant difference between
>> using Red Hat Enterprise Linux and CentOS of the same versions...
>
> There are some subtle differences in configuring networks

Huh? Such as? There shouldn't be...


> and yes I am
> using 3rd party (mysql.com cluster rpms).

That definitely makes a difference vs. using the stock mysql packages,  
but adding those to either RHEL CentOS should be the same difference.


>>> Thank God I live in a state with good breweries !
>>
>> Eh. Stuff is okay over here, I find both Washington and California
>> breweries better. :)
>
> well, I'm also swayed by the fact that on the east coast there is much
> greater access to UK (and especially Scottish) ale, and given that I  
> am
> an ex-pat Scot, that works for me :)

Heh, to each his (or her) own. :)



>>> On Tue, 2009-09-08 at 12:06 -0400, theBlueSage wrote:
>>>> Hi All,
>>>>
>>>> I trying to configure heartbeat-ldirectord on a pair of Centos  
>>>> boxes
>>>> that form part of a mysql cluster. The cluster works and is totally
>>>> happy. I am just trying to load balance the two SQL servers (APIs)
>>>>
>>>> In the 'debian way' I can add the following line to a network
>>>> interface :
>>>>
>>>> #vi /etc/network/interfaces
>>>>
>>>> auto lo:0
>>>> iface lo:0 inet static
>>>> address 10.6.1.75
>>>> netmask 255.255.255.255
>>>> pre-up sysctl -p > /dev/null
>>>>
>>>> however in RedHat/Centos when I edit the equivalent file and add
>>>> the 'pre-up' line :
>>>>
>>>> #vi /etc/sysconfig/network-scripts/ifcfg-lo:0
>>>>
>>>>   DEVICE=lo:0
>>>>   IPADDR=10.6.1.75
>>>>   NETMASK=255.255.255.255
>>>>   NETWORK=10.6.1.0
>>>>   BROADCAST=10.6.1.255
>>>>   ONBOOT=yes
>>>>   pre-up sysctl -p > /dev/null
>>>>
>>>>
>>>> I get a 'command not found' ....
>>>>
>>>> IS there a way to do the pre-up in RH/Centos ?
>>>>
>>>> I also apologize in advance to the RedHat folks for lumping CentOS
>>>> in with them. I have to use
>>>> it at work (at home I am a Debian fan) and although it is 'code
>>>> from the upstream provider' it
>>>> doesnt always act in the same way as 'code from the upstream
>>>> provider'. Also, RedHat docs don't
>>>> always apply to CentOS setups, and CentOs docs dont .... well they
>>>> just dont, really.
>>>>
>>>> in hope,
>>>>
>>>> Richard
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> Discuss mailing list
>>>> Discuss-mNDKBlG2WHs at public.gmane.org
>>>> http://lists.blu.org/mailman/listinfo/discuss
>>>
>>> _______________________________________________
>>> Discuss mailing list
>>> Discuss-mNDKBlG2WHs at public.gmane.org
>>> http://lists.blu.org/mailman/listinfo/discuss
>>
>
> _______________________________________________
> Discuss mailing list
> Discuss-mNDKBlG2WHs at public.gmane.org
> http://lists.blu.org/mailman/listinfo/discuss

-- 
Jarod Wilson
jarod-ajLrJawYSntWk0Htik3J/w at public.gmane.org