-- le script génère les ordres de création sql des tablespaces et datafiles -- si un tablespace contient plusieurs fichier, ceci est pris en compte -- les tablespaces SYSTEM, SYSAUX et UNDO ne sont pas affichés -- -- version avec la taille exacte des datafile occupée sur disque -- -- Historique : -- 2021/11/24 - ajout des tempfiles set head off pages 0 feedback off lines 200 select '------- HOSTNAME : '||host_name||', DB_NAME : '||name||', VERSION : '||version || ' -------' from v$database,v$instance; select '------- Datafiles -------' from dual; SELECT 'CREATE ' || DECODE (ts.bigfile, 'YES', 'BIGFILE ') --assuming smallfile is the default table space || 'TABLESPACE "' || ts.tablespace_name || '" DATAFILE ' || CHR(13) || CHR(10) || LISTAGG(decode(p.value, NULL, ' ''' || df.file_name || '''') || ' SIZE ' || CASE -- si la taille est nulle ou < 1M on retourne 1M WHEN e.used_bytes is NULL or e.used_bytes < (1024*1024) THEN '1M' ELSE to_char(floor(e.used_bytes/(1024*1024))) || 'M' END || DECODE ( df.autoextensible, 'YES', ' AUTOEXTEND ON NEXT ' || ceil(df.increment_by*ts.block_size/1024/1024) || 'M MAXSIZE ' || FLOOR (maxbytes / POWER (1024, 2)) || 'M'), ',' || CHR (13) || CHR (10)) WITHIN GROUP (ORDER BY df.file_id, df.file_name) || ';' ddl FROM dba_tablespaces ts, dba_data_files df, (SELECT file_id, sum(decode(bytes,NULL,0,bytes)) used_bytes FROM dba_extents GROUP by file_id) e, (select VALUE from v$parameter where name='db_create_file_dest') p WHERE ts.tablespace_name not in ('SYSTEM','SYSAUX') and ts.tablespace_name not like '%UNDO%' and e.file_id (+) = df.file_id and ts.tablespace_name = df.tablespace_name GROUP BY ts.tablespace_name, ts.bigfile, ts.block_size ORDER BY ts.tablespace_name; select '------- Tempfiles -------' from dual; SELECT 'CREATE TEMPORARY TABLESPACE "' || ts.tablespace_name || '" TEMPFILE ' || CHR (13) || CHR (10) || LISTAGG(decode(p.value, NULL, ' ''' || df.file_name || '''') || ' SIZE ' || CASE -- si la taille est nulle ou < 1M on retourne 1M WHEN e.used_bytes is NULL or e.used_bytes < (1024*1024) THEN '1M' ELSE to_char(floor(e.used_bytes/(1024*1024))) || 'M' END || DECODE ( df.autoextensible, 'YES', ' AUTOEXTEND ON NEXT ' || ceil(df.increment_by*ts.block_size/1024/1024) || 'M MAXSIZE ' || FLOOR (maxbytes / POWER (1024, 2)) || 'M'), ',' || CHR (13) || CHR (10)) WITHIN GROUP (ORDER BY df.file_id, df.file_name) || ';' ddl FROM dba_tablespaces ts, dba_temp_files df, (SELECT file_id, sum(decode(bytes,NULL,0,bytes)) used_bytes FROM dba_extents GROUP by file_id) e, (select VALUE from v$parameter where name='db_create_file_dest') p WHERE e.file_id (+) = df.file_id and ts.tablespace_name = df.tablespace_name GROUP BY ts.tablespace_name, ts.bigfile, ts.block_size ORDER BY ts.tablespace_name; exit