I ran addmrpt.sql against one of my DB's and got the following output (see below).

My question is how is there a way to tie a SQL_ID (b6usrg82hwsa3) with
a particular SQL statement. Or to simplify, how can I tell what query caused
the issue if the report only returns back a cryptic SQL_ID.

Secondly, is there a rule of thumb to follow as to what percent impact I
should start investigating problems. I would think 53% impact would be
something I need to look into.

Thanks to all who answer


FINDING 1: 53% impact (178 seconds)
-----------------------------------
SQL statements consuming significant database time were found.

RECOMMENDATION 1: SQL Tuning, 43% benefit (144 seconds)
ACTION: Investigate the SQL statement with SQL_ID "b6usrg82hwsa3" for
possible performance improvements.
RELEVANT OBJECT: SQL statement with SQL_ID b6usrg82hwsa3
call dbms_stats.gather_database_stats_job_proc ( )
RATIONALE: SQL statement with SQL_ID "b6usrg82hwsa3" was executed 3
times and had an average elapsed time of 48 seconds.

RECOMMENDATION 2: SQL Tuning, 4.8% benefit (16 seconds)
ACTION: Run SQL Tuning Advisor on the SQL statement with SQL_ID
"8a1pvy4cy8hgv".
RELEVANT OBJECT: SQL statement with SQL_ID 8a1pvy4cy8hgv
insert into histgrm$(obj#,intcol#,row#,bucket,endpoint,col#,epvalue)
values(:1,:2,:3,:4,:5,:6,:7)
ACTION: Investigate the SQL statement with SQL_ID "8a1pvy4cy8hgv" for
possible performance improvements.
RELEVANT OBJECT: SQL statement with SQL_ID 8a1pvy4cy8hgv
insert into histgrm$(obj#,intcol#,row#,bucket,endpoint,col#,epvalue)
values(:1,:2,:3,:4,:5,:6,:7)
RATIONALE: SQL statement with SQL_ID "8a1pvy4cy8hgv" was executed 2715
times and had an average elapsed time of 0.0059 seconds.