• In attendance: WIlliam Dougherty-NI&S, scribe and chair; Lylah Shelor-IT Facilities; Joyce Landreth, Kevin Davis, Nate Smith-UCS;Kimberley Homer-SETI; Karen Herrington-IMS; Wanda Baber, Eliza Lau-Systems Support, Storage Management Team (SMT); Bill deWindt-LT; Russ Fenn-DBAA; Brian Jones-CNS; Susan Brooker-Gross-IT.
  • TSM (backup server) update/upgrade: Liferaft.cc.vt.edu is the current production server.
    Tsmserver1.cc.vt.edu is the new production server that will go production as soon as issues with IBM are resolved. When the new server goes production all users will have to update their dsm.opt to point at the new server. We cannot move all clients overnight. The server can’t handle it. The move will be spread over a month.  Therefore we can’t set up a DNS alias for the old server.
  • Questions being raised include:

1.       Will you be emailing folks before you move them? Yes. SS/SMT has a spreadsheet with 616 clients listed. Team members will send e-mail messages, call, etc. Once a user has updated, they will be blocked on the old machine so a failure would occur if they try to access it.

2.       How many users does this impact? Over 600

3.       Rather than us taking calls about each person moving (or when they break), we think it would be easier to send them a script to automate the change.  Is this something that you can script and make available to both the users and 4Help? TSM admins will interact with UCS staff to work something out.

4.       We would also need a list of users as they are moved.  Is this possible? Yes; these can be posted to the Wiki.

5.       It should also be posted to Techsupport. When the operation is ready to be conducted, lists will be notified.

6.       Why so few users? This is a paying service and another 300+ are on Legato. ($2.40 per GB per year billed monthly)

  • DBAA has the following maintenance scheduled for Sunday, Sept. 12. This has been scheduled for awhile now.
    - From 6am - 10am we will be applying Oracle security patches, a Banner Financial aid and an AR release to the Banner database server Robin. The following databases and services will be unavailable during the maintenance:
    Databases:
    Banner
    vtfbanner
    Services:
    IMS Password Change Application
    Leave Entry And Reporting System
    Make A Gift Today (University Development) 
    Self-Service Banner (Hokie Spa, Hokie Plus, Hokie Team)
    Timetable Of Classes
    Shibboleth (This will take out Google Alumni login for e-mail/A post on the start page is in the works); Will this really be the case or is this old info?
    • Some of these services will be up by 6:30am.
  • From 6am - 8am we will be applying Oracle security patches to the non-Banner server. The production registry and vatech databases will be unavailable. The following services will be affected:
    MyVT
    AAP/ARP
    Budget Tuition
    IMS One Time Password
    Pidgen
    Some apps for College of Agriculture and Life Sciences
    MyVT will be unavailable from 6am - 8am but will also have some of its functionality affected from 8am - 10am.
  • CNS has the following activity in store:* We are continuing upgrades to the university networks as we have been for a few months now. These are being announced on VT-DNET and are coordinated with individual departments residing in the buildings to be upgraded. We also have Cisco engineers on location working with us to solve our wireless centralized controller issues. Steve Lee has been rolling out IOS code upgrades to controllers in a manner that minimizes any user interruptions as bugs are fixed and new updates are released. I have no major announcements for September at this time. We are planning a major upgrade to our core backbone networks in the December time frame. We will be moving from 2xGb port-channeled connectivity between our core nodes to 1x10Gb connections. This will be announced through the VT-DNET  (also available via CSC Dashboard) list when we get closer to the deployment dates for these upgrades (most likely during Christmas break). * Scholar issues are being investigated. Most frequent outages probably based on a bug in the code; hanging connections are sapping resources. Releasing these hung connections is the goal. Five app servers now online. Load was very heavy last night and everything held, so may be seeing light at the end of the tunnel. Using open source code can have this type of issue especially when using an older version.
  • Facilities: By mid-next week, the Data Center wall may begin being cut into (back right side).
  • UCS: We survived rush! The existing CMS system was evaluated as a means of hosting the Computing web site, but "drupal" (drupal.org) was chosen as a better alternative. Migration of data will be done in a staged manner with review of articles as they are moved. Members of this group will be asked to become more involved as process moves forward.
  • Eliza will draft an e-mail message on the TSM migration and send version to UCS for suggestions. 
  • CMS and GIS production databases will also be patched starting at 8am on 9/12/10. Approximately 1/2 hour, but may be longer.
  • SETI-Q&A: 09/09/10 at 7am; new "Student Level Codes" will be loaded. This will only affect active students; will not be retroactive. This will be pushed out for use in issuing e-tokens to grad students. Other uses will no doubt be found later. Access to the attribute must be requested by Enterprise Directory Services users and The new attribute will be available to subscribers of ED-ID services who specifically request it, which will probably only be TAS 3.0 in the beginning.  This will allow TAS 3.0 to grant Personal Digital Certificates on eTokens to ECE graduate students.  There will be no changes to the DAT for the foreseeable future.
  • DCSS is scheduled for October 20th!