CSCI 679 – DBA WORKSHOP II
Spring 2008
Recovery Workshop for weeks of
March 25-April 8, 2008:
Student teams consisting of 3 to 4 persons each will be formed. One target machine will be identified per student team. For each recovery scenario that is undertaken, there will be a project leader chosen to direct that recovery effort. Each student on the team will serve at least once as a project lead, thus at least 3 (or 4) scenarios will be completed. The project leader is responsible for recording the team’s efforts in the following 5 steps and for providing supplementary evidence from the database to support the team’s diagnosis and successful recovery. Be sure to check out all available resources and documentation. Do not count on one source as being completely accurate. Some of your course materials may contain inaccuracies due to attempts to edit old versions of the notes that were used in earlier releases of Oracle.
Here are the teams for the class:
Team One
Shah,Kreya
Belgaonkar,Maneesh V
Deshpande,Anand T
Vohra,Sakina S
Team Two
Koneru,Priyanka
Mohite,Vanitadevi V
Vyas,Amit s
Team Three
Sathrum,Luke Randall
Swisher,Jonathan R
Patil,Ujjwal Uddhav
Lan,Lin
Teams will not be given the same scenarios at the same time. Each will be given a unique order of scenarios to progress through.
The project leader will submit a report for
each recovery scenario they supervise. The report will summarize each of the 5
steps as given below and will comment on the contribution of each team member
to this effort. Here is the 5 step
approach:
1) Record observations
2) Diagnose the problem
3) Formulate a strategy to recover from the problem
4) Execute the strategy
5) Confirm that the problem has been resolved, if not
resolved, iterate steps above until it is resolved.
POSSIBLE RECOVERY SCENARIOS:
Note: a given scenario in the workshop may be
a combination of one or more of the below.
1) Missing one or more system tablespace files
2) Missing one or more (possibly all) copies of the
controlfile
3) Missing a member or possibly all of the members of a
redo logfile group
4) Missing the files or file that comprises the temporary
tablespace
5) Missing a file
or tables from a user’s tablespace
6) Missing the password file
7) Missing the spfile or init.ora file
8) Missing an index file
9) Incomplete recovery due to a missing archive log file
10)
Data in some
tables has been altered by a bad transaction (the transaction was incorrectly
programmed or data was intentionally changed in a way that should not have been
allowed) – how many strategies can be applied here? Flashback versus Incomplete
Recovery
11)
Recovery when
some of the data in a file has been corrupted