Online Redo Log Switch Frequency Map

Posted on May 31, 2019 by Zed DBA Standard 4

A query I find myself often running is the online redo log switch frequency map query, which queries the v$log_history/gv$log_history (for cluster databases) view and show the historical log switch frequency.

Why you might ask? Well it’s important to see how frequent log switches are occurring as Oracle’s rule of thumb is to not switch more then 3 logs per hour (20 minutes of redo) at peak DML activity to prevent excessive checkpoints. The query can also highlight abnormal churn (DML activity).

Online Redo Log Switch Frequency Map Query

set pages 999 lines 400
col h0 format 999
col h1 format 999
col h2 format 999
col h3 format 999
col h4 format 999
col h5 format 999
col h6 format 999
col h7 format 999
col h8 format 999
col h9 format 999
col h10 format 999
col h11 format 999
col h12 format 999
col h13 format 999
col h14 format 999
col h15 format 999
col h16 format 999
col h17 format 999
col h18 format 999
col h19 format 999
col h20 format 999
col h21 format 999
col h22 format 999
col h23 format 999
SELECT TRUNC (first_time) "Date", inst_id, TO_CHAR (first_time, 'Dy') "Day",
 COUNT (1) "Total",
 SUM (DECODE (TO_CHAR (first_time, 'hh24'), '00', 1, 0)) "h0",
 SUM (DECODE (TO_CHAR (first_time, 'hh24'), '01', 1, 0)) "h1",
 SUM (DECODE (TO_CHAR (first_time, 'hh24'), '02', 1, 0)) "h2",
 SUM (DECODE (TO_CHAR (first_time, 'hh24'), '03', 1, 0)) "h3",
 SUM (DECODE (TO_CHAR (first_time, 'hh24'), '04', 1, 0)) "h4",
 SUM (DECODE (TO_CHAR (first_time, 'hh24'), '05', 1, 0)) "h5",
 SUM (DECODE (TO_CHAR (first_time, 'hh24'), '06', 1, 0)) "h6",
 SUM (DECODE (TO_CHAR (first_time, 'hh24'), '07', 1, 0)) "h7",
 SUM (DECODE (TO_CHAR (first_time, 'hh24'), '08', 1, 0)) "h8",
 SUM (DECODE (TO_CHAR (first_time, 'hh24'), '09', 1, 0)) "h9",
 SUM (DECODE (TO_CHAR (first_time, 'hh24'), '10', 1, 0)) "h10",
 SUM (DECODE (TO_CHAR (first_time, 'hh24'), '11', 1, 0)) "h11",
 SUM (DECODE (TO_CHAR (first_time, 'hh24'), '12', 1, 0)) "h12",
 SUM (DECODE (TO_CHAR (first_time, 'hh24'), '13', 1, 0)) "h13",
 SUM (DECODE (TO_CHAR (first_time, 'hh24'), '14', 1, 0)) "h14",
 SUM (DECODE (TO_CHAR (first_time, 'hh24'), '15', 1, 0)) "h15",
 SUM (DECODE (TO_CHAR (first_time, 'hh24'), '16', 1, 0)) "h16",
 SUM (DECODE (TO_CHAR (first_time, 'hh24'), '17', 1, 0)) "h17",
 SUM (DECODE (TO_CHAR (first_time, 'hh24'), '18', 1, 0)) "h18",
 SUM (DECODE (TO_CHAR (first_time, 'hh24'), '19', 1, 0)) "h19",
 SUM (DECODE (TO_CHAR (first_time, 'hh24'), '20', 1, 0)) "h20",
 SUM (DECODE (TO_CHAR (first_time, 'hh24'), '21', 1, 0)) "h21",
 SUM (DECODE (TO_CHAR (first_time, 'hh24'), '22', 1, 0)) "h22",
 SUM (DECODE (TO_CHAR (first_time, 'hh24'), '23', 1, 0)) "h23",
 ROUND (COUNT (1) / 24, 2) "Avg"
FROM gv$log_history
WHERE thread# = inst_id
AND first_time > sysdate -7
GROUP BY TRUNC (first_time), inst_id, TO_CHAR (first_time, 'Dy')
ORDER BY 1,2;

Online Redo Log Switch Frequency Map Output

SQL> set pages 999 lines 400
SQL> col h0 format 999
SQL> col h1 format 999
SQL> col h2 format 999
SQL> col h3 format 999
SQL> col h4 format 999
SQL> col h5 format 999
SQL> col h6 format 999
SQL> col h7 format 999
SQL> col h8 format 999
SQL> col h9 format 999
SQL> col h10 format 999
SQL> col h11 format 999
SQL> col h12 format 999
SQL> col h13 format 999
SQL> col h14 format 999
SQL> col h15 format 999
SQL> col h16 format 999
SQL> col h17 format 999
SQL> col h18 format 999
SQL> col h19 format 999
SQL> col h20 format 999
SQL> col h21 format 999
SQL> col h22 format 999
SQL> col h23 format 999
SQL> SELECT TRUNC (first_time) "Date", inst_id, TO_CHAR (first_time, 'Dy') "Day",
SQL>  COUNT (1) "Total",
SQL>  SUM (DECODE (TO_CHAR (first_time, 'hh24'), '00', 1, 0)) "h0",
SQL>  SUM (DECODE (TO_CHAR (first_time, 'hh24'), '01', 1, 0)) "h1",
SQL>  SUM (DECODE (TO_CHAR (first_time, 'hh24'), '02', 1, 0)) "h2",
SQL>  SUM (DECODE (TO_CHAR (first_time, 'hh24'), '03', 1, 0)) "h3",
SQL>  SUM (DECODE (TO_CHAR (first_time, 'hh24'), '04', 1, 0)) "h4",
SQL>  SUM (DECODE (TO_CHAR (first_time, 'hh24'), '05', 1, 0)) "h5",
SQL>  SUM (DECODE (TO_CHAR (first_time, 'hh24'), '06', 1, 0)) "h6",
SQL>  SUM (DECODE (TO_CHAR (first_time, 'hh24'), '07', 1, 0)) "h7",
SQL>  SUM (DECODE (TO_CHAR (first_time, 'hh24'), '08', 1, 0)) "h8",
SQL>  SUM (DECODE (TO_CHAR (first_time, 'hh24'), '09', 1, 0)) "h9",
SQL>  SUM (DECODE (TO_CHAR (first_time, 'hh24'), '10', 1, 0)) "h10",
SQL>  SUM (DECODE (TO_CHAR (first_time, 'hh24'), '11', 1, 0)) "h11",
SQL>  SUM (DECODE (TO_CHAR (first_time, 'hh24'), '12', 1, 0)) "h12",
SQL>  SUM (DECODE (TO_CHAR (first_time, 'hh24'), '13', 1, 0)) "h13",
SQL>  SUM (DECODE (TO_CHAR (first_time, 'hh24'), '14', 1, 0)) "h14",
SQL>  SUM (DECODE (TO_CHAR (first_time, 'hh24'), '15', 1, 0)) "h15",
SQL>  SUM (DECODE (TO_CHAR (first_time, 'hh24'), '16', 1, 0)) "h16",
SQL>  SUM (DECODE (TO_CHAR (first_time, 'hh24'), '17', 1, 0)) "h17",
SQL>  SUM (DECODE (TO_CHAR (first_time, 'hh24'), '18', 1, 0)) "h18",
SQL>  SUM (DECODE (TO_CHAR (first_time, 'hh24'), '19', 1, 0)) "h19",
SQL>  SUM (DECODE (TO_CHAR (first_time, 'hh24'), '20', 1, 0)) "h20",
SQL>  SUM (DECODE (TO_CHAR (first_time, 'hh24'), '21', 1, 0)) "h21",
SQL>  SUM (DECODE (TO_CHAR (first_time, 'hh24'), '22', 1, 0)) "h22",
SQL>  SUM (DECODE (TO_CHAR (first_time, 'hh24'), '23', 1, 0)) "h23",
SQL>  ROUND (COUNT (1) / 24, 2) "Avg"
SQL> FROM gv$log_history
SQL> WHERE thread# = inst_id
SQL> AND first_time > sysdate -7
SQL> GROUP BY TRUNC (first_time), inst_id, TO_CHAR (first_time, 'Dy')
SQL> ORDER BY 1,2;

Date         INST_ID Day      Total   h0   h1   h2   h3   h4   h5   h6   h7   h8   h9  h10  h11  h12  h13  h14  h15  h16  h17  h18  h19  h20  h21  h22  h23        Avg
--------- ---------- --- ---------- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ----------
24-MAY-19          1 Fri          1    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0    1        .04
24-MAY-19          2 Fri          1    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0    1        .04
25-MAY-19          1 Sat         56    1    1    3    0    2    2    0    7    0    0   10    4    4    6    3    1    2    4    1    1    1    2    0    1       2.33
25-MAY-19          2 Sat         62    0    3    3    0    1    2    0    7    0    0    8    5    2    2    3    5    4    2    3    3    5    4    0    0       2.58
26-MAY-19          1 Sun         56    1    1    3    0    2    5    0   11    0    4    1    1    7    7    3    5    1    0    0    0    1    1    1    1       2.33
26-MAY-19          2 Sun         28    1    3    3    0    1    4    0    5    0    1    1    0    2    3    1    1    1    0    0    0    0    0    1    0       1.17
27-MAY-19          1 Mon         33    1    1    3    0    1    4    0    6    5    1    0    1    1    2    1    1    1    0    0    0    1    1    0    2       1.38
27-MAY-19          2 Mon         21    0    3    3    0    0    4    0    3    2    0    0    1    0    1    0    1    0    1    0    0    0    0    1    1        .88
28-MAY-19          1 Tue         43    2    3    3    0    3    5    1    4    3    1    3    0    1    1    1    1    2    3    2    1    1    0    1    1       1.79
28-MAY-19          2 Tue         38    2    4    3    1    1    4    1    7    1    1    1    1    0    1    0    0    2    4    0    1    0    1    1    1       1.58
29-MAY-19          1 Wed         58    2    4    4    1    3    5    2   12    3    1    3    1    2    5    4    1    1    0    1    0    1    0    2    0       2.42
29-MAY-19          2 Wed         28    0    2    3    0    1    4    1    5    1    0    1    1    0    2    1    1    1    1    0    0    1    1    0    1       1.17
30-MAY-19          1 Thu         39    2    3    3    0    1    5    1    7    1    1    2    3    1    2    2    1    0    1    1    0    0    1    0    1       1.63
30-MAY-19          2 Thu         29    0    1    3    0    2    4    1    4    2    1    1    1    1    1    2    0    1    2    0    0    1    1    0    0       1.21
31-MAY-19          1 Fri        153    2    1    3    0    2    4    1    8    3    0   23   29   10    2    1   18   20    6   12    2    3    3    0    0       6.38
31-MAY-19          2 Fri        223    1    2    3    0    1    4    0    3    1    0   71   86   29    1    0    6    7    2    4    0    1    1    0    0       9.29

16 rows selected.

SQL>

Where the log switch frequency exceeds 3 per hour, I have highlighted as red.

From the above output, we can see that most the time the 3 log switch per hour is met, with the 5th and 7th hour exceeding the threshold. Also we can see today there as been abnormal and excessive churn at 10th and 11th hour.

Investigating this further, we look at v$log to see the size of the online redo logs:

SQL> set pages 999 lines 400
SQL> col FIRST_CHANGE# format 999999999999999
SQL> select GROUP#, THREAD#, SEQUENCE#, BYTES/1024/1024 SIZE_MB, BLOCKSIZE, MEMBERS, ARCHIVED, STATUS, FIRST_CHANGE#, FIRST_TIME, NEXT_CHANGE#, NEXT_TIME, CON_ID from v$log;

    GROUP#    THREAD#  SEQUENCE#    SIZE_MB  BLOCKSIZE    MEMBERS ARC STATUS              FIRST_CHANGE# FIRST_TIM NEXT_CHANGE# NEXT_TIME     CON_ID
---------- ---------- ---------- ---------- ---------- ---------- --- ---------------- ---------------- --------- ------------ --------- ----------
         1          1      40649       4096        512          2 YES INACTIVE             137921001754 31-MAY-19   1.3793E+11 31-MAY-19          0
         2          1      40650       4096        512          2 NO  CURRENT              137930464620 31-MAY-19   2.8147E+14                    0
         3          1      40648       4096        512          2 YES INACTIVE             137920121808 31-MAY-19   1.3792E+11 31-MAY-19          0
         4          2      39319       4096        512          2 NO  CURRENT              137921005288 31-MAY-19   2.8147E+14                    0
         5          2      39317       4096        512          2 YES INACTIVE             137917291807 31-MAY-19   1.3792E+11 31-MAY-19          0
         6          2      39318       4096        512          2 YES INACTIVE             137919186475 31-MAY-19   1.3792E+11 31-MAY-19          0

6 rows selected.

SQL>

As we can see the online redo logs are 4GB, which is the default size on Oracle Exadata, which this is. We could increase the size of the redo logs but it trade-off between performance verse space usage. The peak we can see was 29 log switches for instance 1 and 86 log switches for instance 2, assuming these were full redo logs before switching, this is 29 + 86 logs switches for the database. Which is 115 for the the hour, equating to a significant 460 GB of churn! But this is an Exadata Machine and is sized and designed for this sort of load 🙂 To resize the redo logs to switch only 3 per hour, we would need to resize to 154GB, this wouldn’t be feasible, so we could either increase to something more reasonable like 10GB or leave as is, depending on what more important, the performance or space usage.

If you found this blog post useful, please like as well as follow me through my various Social Media avenues available on the sidebar and/or subscribe to this oracle blog via WordPress/e-mail.

Thanks

Zed DBA (Zahid Anwar)

Prevent crontab jobs overlapping using flock

Posted on May 31, 2019 by Zed DBA Standard 2

As an Oracle DBA, you may find yourself in the situation where you have crontab jobs overlapping.

For example an RMAN backup takes longer then normal, then overlaps with another RMAN database backup leading to more resources being consumed:

00 00 * * * /home/oracle/scripts/backups/db_backup.sh PROD1 >> /home/oracle/backups/logs/db_backup_PROD1.log 2>&1
00 01 * * * /home/oracle/scripts/backups/db_backup.sh PROD2 >> /home/oracle/backups/logs/db_backup_PROD2.log 2>&1

If the PROD1 backup takes longer then 1 hour, then it will contend with the PROD2 backup when it starts.

Another more recent example for myself, is when I was copying archive logs to AWS for an Oracle Database Standard Edition migration using a manual standby, thus needed to manually transfer archive logs using rsync every 15 minutes:

0,15,30,45 * * * * /home/oracle/copy_Arch_to_AWS.sh > /home/oracle/copy_Arch_to_AWS.log

Ran fine most the time but when there was significant churn in the database, the crontab job would overlap causing several rsync 😦

The easiest solution is to wrap the crontab job in flock 🙂

Using flock

flock is a linux utility that can uses a lock file to determine if the process is already running. The syntax I use is:

flock -x <lockfile> -c '<command>'

The “-x” is to obtain exclude lock and the “-c” is the command to run.

Flock examples

Backup example

00 00 * * * flock -x /home/oracle/scripts/backups/backup.lock -c '/home/oracle/scripts/backups/db_backup.sh PROD1 >> /home/oracle/backups/logs/db_backup_PROD1.log 2>&1'
00 01 * * * flock -x /home/oracle/scripts/backups/backup.lock -c '/home/oracle/scripts/backups/db_backup.sh PROD2 >> /home/oracle/backups/logs/db_backup_PROD2.log 2>&1'

Now when the backup for PROD2 starts flock will check for the lock and will see if exist and will not run the command until the backup for PROD1 is completed 🙂

Archive log copy example

0,15,30,45 * * * * flock -x /home/oracle/copy_Arch_to_AWS.lock -c '/home/oracle/copy_Arch_to_AWS.sh > /home/oracle/copy_Arch_to_AWS.log'

Now when the job runs, an exclusive lock is taken an hence when it runs again in 15 minutes if there an existing run, then it will not run the command until the previous one is completed 🙂 This will essentially queue the copies instead of them overlapping causing several rsync, which just exacerbate the issue.

Advance use of flock

Timeout

You can add “-w <seconds>“, which is the amount of time to wait for exclusive lock before exiting without running command. for example:

0,15,30,45 * * * * flock -w 300 -x /home/oracle/copy_Arch_to_AWS.lock -c '/home/oracle/copy_Arch_to_AWS.sh > /home/oracle/copy_Arch_to_AWS.log'

Now flock will wait 5 minutes for the previous archive log copy job to complete before exiting without running the command for that run 🙂

Viewing lock

If you want to see the lock taken by flock, you can run :

[oracle@dc1sbxdb001 ~]$ fuser -v /home/oracle/copy_Arch_to_AWS.lock
                                    USER   PID    ACCESS COMMAND
/home/oracle/copy_Arch_to_AWS.lock: oracle 341039 f....  flock
                                    oracle 341040 f....  rsync

If you found this blog post useful, please like as well as follow me through my various Social Media avenues available on the sidebar and/or subscribe to this oracle blog via WordPress/e-mail.

Thanks

Zed DBA (Zahid Anwar)

Adding a Datafile to Temp Tablespace

Posted on April 10, 2018 by Zed DBA Standard 2

When monitoring Tablespace Usage (see my Tablespace Usage blog post for more info), there comes a point when you need to add a datafile to the temp tablespace to allow for growth and more importantly get below a monitoring threshold for example in OEM or OpsView.

Query to see Current Temp Datafiles State

To see the current state of the temp datafiles:

set pages 999
set lines 400
col FILE_NAME format a75
select d.TABLESPACE_NAME, d.FILE_NAME, d.BYTES/1024/1024 SIZE_MB, d.AUTOEXTENSIBLE, d.MAXBYTES/1024/1024 MAXSIZE_MB, d.INCREMENT_BY*(v.BLOCK_SIZE/1024)/1024 INCREMENT_BY_MB
from dba_temp_files d,
 v$tempfile v
where d.FILE_ID = v.FILE#
order by d.TABLESPACE_NAME, d.FILE_NAME;

Please Note: For pluggable databases, please ensure you are in the right PDB as temp datafiles can be at container level and/or pluggable database level.

Output from Current Temp Datafiles State

This is the output you will get from the query:

TABLESPACE_NAME FILE_NAME SIZE_MB AUT MAXSIZE_MB INCREMENT_BY_MB
------------------------------ --------------------------------------------------------------------------- ---------- --- ---------- ---------------
TEMP +DATAC1/VERS/TEMPFILE/temp.451.891367325 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.452.891367321 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.454.891367305 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.455.891367301 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.457.891367299 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.458.891367295 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.459.891367159 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.461.891367165 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.463.891367201 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.465.891367203 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.466.891367207 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.468.891367223 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.469.891367227 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.471.891367243 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.472.891367251 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.473.891367255 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.475.891367261 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.477.891367289 32767 YES 32767 1024
...

38 rows selected.

SQL>

The … represents the several lines removed to make the output readable 🙂

Add Temp Datafile to Temp Tablespace

To add a temp datafile to ‘TEMP‘ to be initially 10G, auto extendable by 1G to maxsize of 32Gb:

ALTER TABLESPACE TEMP ADD TEMPFILE '+DATAC1' SIZE 10G AUTOEXTEND ON NEXT 1G MAXSIZE 32767M;

See Oracle Documentation for more info and Syntax:
https://docs.oracle.com/database/121/SQLRF/statements_3002.htm

Once added you’ll see the temp datafile in the above query:

TABLESPACE_NAME FILE_NAME SIZE_MB AUT MAXSIZE_MB INCREMENT_BY_MB
------------------------------ --------------------------------------------------------------------------- ---------- --- ---------- ---------------
TEMP +DATAC1/VERS/TEMPFILE/temp.451.891367325 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.452.891367321 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.454.891367305 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.455.891367301 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.457.891367299 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.458.891367295 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.459.891367159 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.461.891367165 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.463.891367201 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.465.891367203 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.466.891367207 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.468.891367223 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.469.891367227 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.471.891367243 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.472.891367251 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.473.891367255 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.475.891367261 32767 YES 32767 1024
TEMP +DATAC1/VERS/TEMPFILE/temp.477.891367289 32767 YES 32767 1024
...
TEMP +DATAC1/VERS/TEMPFILE/temp.843.973079825 10240 YES 32767 1024

39 rows selected.

SQL>

Related Blog Posts:
Querying Tablespace Usage
Adding a Datafile to Tablespace

Updates
27th April 2018: Change the query to use v$tempfile instead of v$datafile as doesn’t work for PDBs.

If you found this blog post useful, please like as well as follow me through my various Social Media avenues available on the sidebar and/or subscribe to this oracle blog via WordPress/e-mail.

Thanks

Zed DBA (Zahid Anwar)

How to fix queries on DBA_FREE_SPACE that are slow

Posted on January 25, 2018 by Zed DBA Standard 15

I found myself in a situation where OpsView, a monitoring tool, was having difficulty monitoring the tablespaces for a particular pluggable database.

Upon investigation it was found the queries against the dictionary table DBA_FREE_SPACE were taking a very long time:

SQL> set timing on
SQL> select nvl(sum(dfs.bytes)/1024/1024,0) from dba_free_space dfs where dfs.tablespace_name = 'USERS';

NVL(SUM(DFS.BYTES)/1024/1024,0)
-------------------------------
 70.75

Elapsed: 00:00:10.98

There are 60 tablespaces in this pluggable database, which the time varied querying each tablespace, but was by far where most the time was spent.

I wrote a PL/SQL block to mimic Opsview as I didn’t want to create an object (procedure) in this customer’s database:

SET SERVEROUTPUT ON
SET TIMING ON
DECLARE
 cursor ts_names is select tablespace_name from dba_tablespaces where contents != 'TEMPORARY';
 sql_used VARCHAR(200) := 'select nvl(sum(dfs.bytes)/1024/1024,0) from dba_free_space dfs where dfs.tablespace_name = ''text_string''';
 sql_free VARCHAR(200) := 'select nvl(sum(dfs.bytes)/1024/1024,0) from dba_free_space dfs where dfs.tablespace_name = ''text_string''';
 sql_max VARCHAR(200) := 'select sum(maxbytes-bytes)/1024/1024 from dba_data_files where AUTOEXTENSIBLE = ''YES'' and maxbytes>bytes and tablespace_name = ''text_string''';
 num_out NUMBER;
BEGIN
 FOR ts_name in ts_names
 LOOP
 --sql for used space
 EXECUTE IMMEDIATE replace(sql_used, 'text_string', ts_name.tablespace_name) into num_out;
 dbms_output.put_line(replace(sql_used, 'text_string', ts_name.tablespace_name));
 dbms_output.put_line(num_out);
 --sql for free space
 EXECUTE IMMEDIATE replace(sql_free, 'text_string', ts_name.tablespace_name) into num_out;
 dbms_output.put_line(replace(sql_free, 'text_string', ts_name.tablespace_name));
 dbms_output.put_line(num_out);
 --sql for max
 EXECUTE IMMEDIATE replace(sql_max, 'text_string', ts_name.tablespace_name) into num_out;
 dbms_output.put_line(replace(sql_max, 'text_string', ts_name.tablespace_name));
 dbms_output.put_line(num_out);
 END LOOP;
END;
/

I ran this and the total time was shocking 😐 :

SQL> --SET SERVEROUTPUT ON
SQL> SET TIMING ON
SQL> DECLARE
 2 cursor ts_names is select tablespace_name from dba_tablespaces where contents != 'TEMPORARY';
 3 sql_used VARCHAR(200) := 'select nvl(sum(dfs.bytes)/1024/1024,0) from dba_free_space dfs where dfs.tablespace_name = ''text_string''';
 4 sql_free VARCHAR(200) := 'select nvl(sum(dfs.bytes)/1024/1024,0) from dba_free_space dfs where dfs.tablespace_name = ''text_string''';
 5 sql_max VARCHAR(200) := 'select sum(maxbytes-bytes)/1024/1024 from dba_data_files where AUTOEXTENSIBLE = ''YES'' and maxbytes>bytes and tablespace_name = ''text_string''';
 6 num_out NUMBER;
 7 BEGIN
 8 FOR ts_name in ts_names
 9 LOOP
 10 --sql for used space
 11 EXECUTE IMMEDIATE replace(sql_used, 'text_string', ts_name.tablespace_name) into num_out;
 12 dbms_output.put_line(replace(sql_used, 'text_string', ts_name.tablespace_name));
 13 dbms_output.put_line(num_out);
 14 --sql for free space
 15 EXECUTE IMMEDIATE replace(sql_free, 'text_string', ts_name.tablespace_name) into num_out;
 16 dbms_output.put_line(replace(sql_free, 'text_string', ts_name.tablespace_name));
 17 dbms_output.put_line(num_out);
 --sql for max
 18 19 EXECUTE IMMEDIATE replace(sql_max, 'text_string', ts_name.tablespace_name) into num_out;
 20 dbms_output.put_line(replace(sql_max, 'text_string', ts_name.tablespace_name));
 21 dbms_output.put_line(num_out);
 22 END LOOP;
END;
 23 24 /

PL/SQL procedure successfully completed.

Elapsed: 00:21:30.94
SQL>

So I searched My Oracle Support (MOS) and found the following MOS note:
Queries on DBA_FREE_SPACE are Slow (Doc ID 271169.1)

Which states:
“1) In release 10g, the view dba_free_space was modified to also include objects in the recycle bin.

2) Large number of objects in the recyclebin can slow down queries on dba_free_space.

3) This is a normal behaviour.

4) For release 11g, the view dba_free_space doesn’t contain a hint which in case when there is only few objects in recyclebin, you may want to gather underlying stats of tables/dictionary to get better performance.”

The database indeed did have a lot of objects in the recycle bin (in the pluggable database):

SQL> SELECT count(*) from dba_recyclebin;

 COUNT(*)
----------
 27615

SQL>

With most of them drop recently:

SQL> select trunc(to_date(DROPTIME,'YYYY-MM-DD:HH24:MI:SS')), count(*) from dba_recyclebin group by trunc(to_date(DROPTIME,'YYYY-MM-DD:HH24:MI:SS'))
  2  order by 1
  3 /

TRUNC(TO_ COUNT(*)
--------- ----------
24-SEP-16 2
...
19-JAN-18 2506
20-JAN-18 4322
21-JAN-18 4321
22-JAN-18 4320
23-JAN-18 4321
24-JAN-18 4321
25-JAN-18 2446

421 rows selected.

SQL>

So I purged the recycle bin (with customers permission) and re-ran the check:

SQL> purge dba_recyclebin;

DBA Recyclebin purged.

Elapsed: 00:06:30.39
SQL> --SET SERVEROUTPUT ON
SET TIMING ON
SQL> SQL> DECLARE
 2 cursor ts_names is select tablespace_name from dba_tablespaces where contents != 'TEMPORARY';
 3 sql_used VARCHAR(200) := 'select nvl(sum(dfs.bytes)/1024/1024,0) from dba_free_space dfs where dfs.tablespace_name = ''text_string''';
 4 sql_free VARCHAR(200) := 'select nvl(sum(dfs.bytes)/1024/1024,0) from dba_free_space dfs where dfs.tablespace_name = ''text_string''';
 5 sql_max VARCHAR(200) := 'select sum(maxbytes-bytes)/1024/1024 from dba_data_files where AUTOEXTENSIBLE = ''YES'' and maxbytes>bytes and tablespace_name = ''text_string''';
 6 num_out NUMBER;
 7 BEGIN
 8 FOR ts_name in ts_names
 9 LOOP
 10 --sql for used space
 11 EXECUTE IMMEDIATE replace(sql_used, 'text_string', ts_name.tablespace_name) into num_out;
 12 dbms_output.put_line(replace(sql_used, 'text_string', ts_name.tablespace_name));
 13 dbms_output.put_line(num_out);
 14 --sql for free space
 15 EXECUTE IMMEDIATE replace(sql_free, 'text_string', ts_name.tablespace_name) into num_out;
 16 dbms_output.put_line(replace(sql_free, 'text_string', ts_name.tablespace_name));
 17 dbms_output.put_line(num_out);
 18 --sql for max
 19 EXECUTE IMMEDIATE replace(sql_max, 'text_string', ts_name.tablespace_name) into num_out;
 20 dbms_output.put_line(replace(sql_max, 'text_string', ts_name.tablespace_name));
 21 dbms_output.put_line(num_out);
 22 END LOOP;
 23 END;
 24 /

PL/SQL procedure successfully completed.

Elapsed: 00:02:46.25
SQL>

Result, the duration of the PL/SQL block went from 21 minutes to just under 3 minutes. However I need it to go under 2 minutes as this was the timeout for OpsView.

So I proceed with the next recommendation in the MOS note of gather dictionary and fixed table stats (with customers permission) using MOS note:
How to Gather Statistics on Objects Owned by the ‘SYS’ User and ‘Fixed’ Objects (Doc ID 457926.1)

SQL> EXEC DBMS_STATS.GATHER_DICTIONARY_STATS;

PL/SQL procedure successfully completed.

Elapsed: 00:00:20.49

SQL> EXEC DBMS_STATS.GATHER_FIXED_OBJECTS_STATS;

PL/SQL procedure successfully completed.

Elapsed: 00:04:28.07

SQL> --SET SERVEROUTPUT ON
SET TIMING ON
SQL> SQL> DECLARE
 2 cursor ts_names is select tablespace_name from dba_tablespaces where contents != 'TEMPORARY';
 3 sql_used VARCHAR(200) := 'select nvl(sum(dfs.bytes)/1024/1024,0) from dba_free_space dfs where dfs.tablespace_name = ''text_string''';
 4 sql_free VARCHAR(200) := 'select nvl(sum(dfs.bytes)/1024/1024,0) from dba_free_space dfs where dfs.tablespace_name = ''text_string''';
 5 sql_max VARCHAR(200) := 'select sum(maxbytes-bytes)/1024/1024 from dba_data_files where AUTOEXTENSIBLE = ''YES'' and maxbytes>bytes and tablespace_name = ''text_string''';
 6 num_out NUMBER;
 7 BEGIN
 8 FOR ts_name in ts_names
 9 LOOP
 10 --sql for used space
 11 EXECUTE IMMEDIATE replace(sql_used, 'text_string', ts_name.tablespace_name) into num_out;
 12 dbms_output.put_line(replace(sql_used, 'text_string', ts_name.tablespace_name));
 13 dbms_output.put_line(num_out);
 14 --sql for free space
 15 EXECUTE IMMEDIATE replace(sql_free, 'text_string', ts_name.tablespace_name) into num_out;
 16 dbms_output.put_line(replace(sql_free, 'text_string', ts_name.tablespace_name));
 17 dbms_output.put_line(num_out);
 18 --sql for max
 19 EXECUTE IMMEDIATE replace(sql_max, 'text_string', ts_name.tablespace_name) into num_out;
 20 dbms_output.put_line(replace(sql_max, 'text_string', ts_name.tablespace_name));
 21 dbms_output.put_line(num_out);
 22 END LOOP;
 23 END;
 24 /

PL/SQL procedure successfully completed.

Elapsed: 00:00:04.53
SQL>

Bingo! the duration of the PL/SQL block went down to 4 seconds 🙂

PLEASE NOTE: This still effects non-pluggable databases, however in pluggable databases, you need to purge the recycle bin for where the dropped objects are, the container database and the pluggable databases require independent purge.

If you found this blog post useful, please like as well as follow me through my various Social Media avenues available on the sidebar and/or subscribe to this oracle blog via WordPress/e-mail.

Thanks

Zed DBA (Zahid Anwar)

New READ Object Privilege in 12cR1

Posted on November 20, 2017 by Zed DBA Standard 1

In writing a blog post about:
Creating a Read Only Database User Account in an Oracle Database

It came to my attention of the new “READ” object privilege, which is a New Feature in 12.1.0.2:
Changes in Oracle Database 12c Release 1 (12.1.0.2)
READ and SELECT Object Privileges

The “SELECT” object privilege in addition to querying the table, allows the user to:
LOCK TABLE table_name IN EXCLUSIVE MODE;
SELECT … FROM table_name FOR UPDATE;

The New Feature of “READ” object privilege, does not allow the user to lock tables in exclusive mode nor select table for update.

Prior to 12.1.0.2, the “SELECT” object privilege is only available which allows the locking:

GRANT SELECT ON ... TO ...;

12.1.0.2 onwards, the new “READ” object privilege is available which doesn’t allow the locking:

GRANT READ ON ... TO ...;

This also applies to the “SELECT ON ANY TABLE“, prior to 12.1.0.2 which allows the locking:

GRANT SELECT ANY TABLE TO ...;

12.1.0.2 onwards, the new “READ ON ANY TABLE” object privilege is available which doesn’t allow the locking:

GRANT READ ANY TABLE TO ...;

If you found this blog post useful, please like as well as follow me through my various Social Media avenues available on the sidebar and/or subscribe to this oracle blog via WordPress/e-mail.

Thanks

Zed DBA (Zahid Anwar)

Zed DBA's Oracle Blog

The Oracle DBA Specialist – OCM, OCP, OCE, OCS, ACE

Tag Archives: Database Administration

Online Redo Log Switch Frequency Map

Online Redo Log Switch Frequency Map Query

Online Redo Log Switch Frequency Map Output

Prevent crontab jobs overlapping using flock

Using flock

Flock examples

Backup example

Archive log copy example

Advance use of flock

Timeout

Viewing lock

Adding a Datafile to Temp Tablespace

Query to see Current Temp Datafiles State

Output from Current Temp Datafiles State

Add Temp Datafile to Temp Tablespace

How to fix queries on DBA_FREE_SPACE that are slow

New READ Object Privilege in 12cR1