Michael Olson wrote:
> I'm not familiar with either BIND or webalizer, but if no one else
> steps up, I'll take them on.
>   
>> Whoever signs up for BIND should set our BINDs up for local DNS
>> resolutions.
>>     
>
> Which I don't know how to do, so it really would be best if someone
> who did know how to work with BIND took it over.
>
>   

Based on past experience, it seems that there aren't many people in the 
group who are drooling over the possibility of BIND administration, so I 
wanted to comment on this setup in case I can be of any help, having 
configured a couple of BIND servers before.

There are two ways in which Adam's comment about the "local resolution" 
can be interpreted.  First, mire and deleuze should probably be 
configured as "caching nameservers," along the lines of the following 
TLDP page:

http://tldp.org/HOWTO/DNS-HOWTO-3.html

Then you will have to make sure that all client machines on the local 
net have the ip addresses for both of our nameservers in their 
/etc/resolv.conf file.  This will make sure that only the first lookup 
of a hostname or ip address leaves our network.  Subsequent requests 
should be cached, thus reducing traffic on the internet connection that 
we have and speeding up host resolution significantly.

Second, our name servers should be responsible for hcoop.net hosts.  
However, it seems that this part should be handled by domtool and no 
configuration should need to be done on your part.  That is, unless 
hcoop.net is seen as a "special case" in terms of DNS for some reason.  
Adam could probably clarify this.

If you're really uneasy about doing this, and no one else wants the job, 
I wouldn't mind coming back as an admin in a limited capacity, perhaps 
as a "name resolution admin."  Related to this, I wrote a quick program 
before that would parse the old domtool text files and tell users about 
where possible errors were that would result in a crippled DNS setup for 
their domains.  I'd like to adopt this to the domtool2 program and make 
it available if there is any interest.

http://justin.phq.org/software/

Of course, my feelings wouldn't be hurt if the board doesn't want me to 
come back in just this limited capacity.  I just figured I'd offer, 
especially since there seemed to be no interest in such a job before 
among the current admins, and I want to help get the new systems up and 
running as quickly as possible.

Finally, I just looked at the wiki pages and didn't see any mention of 
monitoring tools being set up for service reliability.  When I signed up 
as a volunteer admin before, I mentioned that I would be interested in 
setting up nagios to monitor critical areas of the network.  Is anyone 
willing to take this on?  If not, I'd be willing to see what I can do to 
set up some system monitoring tools that would alert a group of admins 
of potential problems (including myself, when I'm able to jump in and 
troubleshoot something).  I could also work on extending this to some 
dell monitoring utilities.

Anyway, I know that this would be departing somewhat from the idea of 
"three volunteer admins," but I wanted to see if I could be of any use 
in getting these new machines up and running as soon as possible.

Either way, thanks again for all of the hard work that everyone has 
done!  It sounds like things are falling into place.

Best,

Justin

_______________________________________________
HCoop-SysAdmin mailing list
[email protected]
http://hcoop.net/cgi-bin/mailman/listinfo/hcoop-sysadmin

Reply via email to