(March 31, 1997)
The purpose of the Service Change Management Policy is to manage changes in a rational and predictable manner so that staff and clients can plan accordingly. Changes require serious forethought, careful monitoring, and followup evaluation to reduce negative impact and increase positive benefits.
Change to immediately restore service to clients.
Change to immediately fix an existing problem.
Notification and documentation are given as soon as possible after the change. In cases where ITS service outage is anticipated, immediate notification is required.
(Examples: Hardware or software failures which disable services.)
Change has major impact on services and clients.
Change is highly visible.
First time this or similar change has been done.
Backout from change is difficult or impossible.
Change itself is complex and/or lengthy.
Notification to both external and internal clients required at least one month in advance of change date and periodically after change. Draft of notification must be reviewed by ITS / Services review group and senior management.
(Examples: removal of production hardware or software; the discontinuance of a service; a major network access change; a major service delivery change.)
Change is significant to clients.
Change is visible to a large group of clients.
Backout from change may require significant effort.
Change itself is difficult, but not necessarily too
involved.
Notification to both external and internal clients required at least two weeks in advance of change date and immediately following change. Draft of notification must be reviewed by ITS /Services review group.
(Examples: documentation changes; new versions or major enhancements of software offerings such as pine and SAS.)
Change has minor impact on clients.
Change is visible to small group of clients.
Change or similar change has been carried out successfully
several times before.
Backout requires moderate effort.
Change itself is easy to enact.
Notification to both external and internal clients required at least one week in advance of change date and immediately following change.
(Examples: temporary hours of operation changes; login script changes.)
Change has negligible impact on clients.
Change is familiar or common to those implementing it.
Change is reliable and known to work.
Change is easy to back out if problems occur.
Notification required at the time of change (either immediately prior to change or immediately after change is made).
(Examples: modem exchanges; making software available for public to download; installing user-supplied but otherwise organizationally unsupported software on central systems.)
Provides additional information concerning an on-going problem or change.
"ITS_Changes@unc.edu".
If SIGNIFICANT change, send draft of notification to review group. Review suggestions. Send notice to
"ITS_Changes@unc.edu".
If MINOR, TRANSPARENT, FOLLOWUP or EMERGENCY change, send notice directly to
"ITS_Changes@unc.edu".
Effect the change according to the scheduled date.
Date: Wed, 02 Feb 94 14:20 EST To: ITS_Changes From: Change Owner ID@NODE.OIT.UNC.EDU Subject: MINOR: Notice of IBM Systems change on Wednesday Feb 9 Change Category: MINOR On Wednesday morning 02/09/94, we will be putting into production a change in the routing for TSO (MVS) mail. On the IBM mainframe, all outgoing mail from TSO (MVS) will be routed via a new software process, SMTP. This change will allow all internet and local (UNC) mail to be routed directly instead of passing it through VM. The installation will occur during system test time (5:30 a.m. - 8:00 a.m.) and the system will be available by 8:00 a.m. The routing changes will be tested this week during normal test time. This change does not affect user procedures for mail nor does it change any screen layouts.
Date: Wed, 09 Feb 94 08:16 EST To: ITS_Changes From: Change Owner ID@NODE.OIT.UNC.EDU Subject: MINOR: Follow-up: IBM System Change for mail routing Change Category: MINOR As previously posted, a change in the routing for IBM mainframe TSO (MVS) mail went into effect this morning. All outgoing internet mail from MVS is now routed directly through SMTP on MVS instead of being passed to VM for routing. Reminder: This change does not affect user procedures for mail nor does it change any screen layouts.