Orbis 2 Systems Installation, Implementation and Operations Group (SIIOG)
Meeting Notes: February 7, 2002
Present: Gail Barnett, Roy Lechich, Wes Most, Ernie Marinko, Fred Martz, David Gewirtz
Server Implementation Schedule
David Gewirtz joined our meeting for this session to discuss these two issues.
Server Implementation Schedule
Since we received the new Sun Hardware for Orbis 2 in early January, the ITS/Library group has gotten the servers physically installed in their respective racks in the Whitney and Watson data centers, and the production Oracle server (“Lewis”) and Voyager application server (“da Gama”) have been set up and configured with the Solaris operating system, YASSP security, Oracle and the Voyager application, and are ready for the dress rehearsal data migration we will be conducting over the next few weeks. The remaining servers are:
1) “Columbus” – production web
2) “Magellan” – development
3) “Clark” – backup Oracle
4) “de Leone” – backup Voyager
5) “Vespucci” – utility (backup web, media server for tape system, and jumpstart server)
6) “Champlain” – Sun system monitoring
David proposed that, since it should take no more than a few days to set up Vespucci as a jumpstart server, his group set that up next, as this would simplify the setup of the remaining servers. The timeline we discussed is:
- 02/12: Dev Server has Solaris and YASSP
- 02/14: Jumpstart functionality is set up and functional on the utility server.
- 02/14: Production Web Server is configured.
- 02/15: Dev Server finished, with Oracle, Voyager, development environment and user accounts. Oracle will contain the TrainDB database from the epreservation machine which contains the system administration and configuration work done so far by the various implementation groups.
Regarding the remaining servers (Oracle and Voyager backups Clark and de Leone, and system monitoring Champlain), David proposed that, if possible, we leave those available for use by his group as training/testing/tuning platforms until May 3. Since this is brand new hardware and there will be a significant learning curve with the new system, this time would be useful to develop and fine-tune techniques and settings for system operations relating to backup, disaster recovery, failover and system monitoring , as well as SNDR and Instant Image, which we plan to use in order to keep our Oracle databases synchronized.
We agreed that we should be able to obtain a formal response to this request within a week’s time.
David further proposed that we re-visit the issue of operational standards, which as he pointed out
has not been formally visited since Orbis’ inception over a decade ago. Since Orbis 2 will be such a different system from Orbis, with different expectations, these standards need to be re-examined and then re-stated.
A partial list of the standards includes:
- Hours of Availability
- Change management
- Outages for updates, etc.
- Notification for planned outages
- Batch Processing
- Backup Procedures
- Disaster Recovery Procedures
- Security for all servers
David proposed that his group go through these issues and prepare a specifications document for the Library to respond to.
Action Items and Questions
1) Get response to the proposal to hold Clark, de Leone and Champlain for ITS/Library group’s use until May 3.
2) Find out what our immediate backup needs will be until the tape system is finalized. Will the YaleDB database need to be updated daily after it is built during the dress rehearsal data migration?
3) When will the reporting team be formed? David would like to have a representative from his group on this team.