NCSU OIT Integration Services

We haven't had a chance to migrate the content off of this site, so I don't think we'll be docommissoning it on the first of January.    It's good to set specific goals, but it's also good to let them slide when the project isn't on any criticial paths.

 We'll figure out in the new year how we'd like to prioritize integrate.oit's demise. :-)

Folks,

Our newest, hopefully permenant home has been established at http://xteams.oit.ncsu.edu/iso

We haven't been know as "Integration Services" for at least two re-orgs now, and it was either time to move, or allocate some more attention to this site's backend and disaster recovery.

I'm proud that this was the first drupal site in OIT, and I think our successful example paved the way for all the Drupal projects that are now moving forward or on the drawing board.

Folks,

 integrate.oit went off the air for a bit when it got a kernel update.  Just so I'm not the only one who knows the details, here's what I did to bring it back.  I'll print a hardcopy and stick it to my door, and we'll call it DRP. :-)

The integrate site it hosted on a virtual machine, named "drupal" running on wpdev.unity.ad.ncsu.edu .  Yes, the 'dev' means test box that isn't monitored.

You gain access to it with  PuTTY or ssh (regular Unity credentials).  

So as suspected, the issue with folks getting spurious pages about "unix-y" hosts was a mistake of different cultures.

Pager copies cccadm@help.ncsu.edu to get a Remedy call cut for any pages sent for the cccadm on-call group.  The *page* is sent through arch wireless or some other alpha paging system, the e-mail is just a cc so the incident will get a tracking number for later reporting/management.

Remedy is configured to copy the oit-iso-shs@lists.ncsu.edu mailing list.  This is where the unwanted pages at 3am are coming from.

Hi gang,

 Joe and Will seem to have access controls assigned to groups rather then individual people in QIP.  Yay, go team, and woot!

For our monday meeting, I'd like to ask each of you to make a printout of all the VLANS that you have rights to in QIP, and mark up which ones you think the whole group needs, if they are to be responsible even in part for the job you're doing now.  I'd like to compile a map of VLANS, past and present.

So, for what it's worth, the blackberry I've been assigned has unlimited (SMS) text and 600 minutes of talk time a month.

According to the web page at

https://text.vzw.com/customer_site/secure/jsp/overview.jsp

To create a text message, one just e-mails to

mobile_number@vtext.com

Text length is limited to 140 characters. 

 So it turns out that all of the "out of the box" features that nagios provides are available to us on the http://nagios.unity.ncsu.edu site.  It offers you a link to the Sysnews NCSU specific pages, but there's also a link to the raw Nagios Status Page.

 In the left hand pane, you can select a "Service Detail" view that expands all the current service status information for a host group.

Folks,

Since effective today OIT has a new logo, I've poked it onto this site to show our solidarity.

Joe J and I have added most of the servers that he manages certificates for to the monitoring tool at: 

https://sysnews.ncsu.edu/tools-bin/sslkey-status

This is NOT nagios, but a seperate app that will send an e-mail starting 60 days from when a cert expires.   The whole team can add, remove and manage the host lists.

There were several hosts that we couldn't reach from the sysnews servers to monitor them. including:

Partly as a spinoff from last Friday's meeting, and partly from my own reflections on stuff I've done that hasn't worked, I've been wrestling with a question. The question I have is, how do we meet management needs without needless bureaucracy? How can we meet these needs in a sensible, sustainable way?

Roughly, from my experience, management needs: