Category: RAC | Data Guard

Errors and Bugs, GoldenGate, RAC | Data Guard

Failure unregister integrated extract

Some times it’s impossible to unregister Integrated Extract, however it need to exclude to avoid RMAN failures.

Follow below to hack GoldenGate registration:

SQL> select CAPTURE_NAME from dba_capture;

CAPTURE_NAME
------------------------------
OGG$CAP_IE_CAPT

GGSCI (myhost as ggate@foodb) 13> unregister extract IE_CAPT database

ERROR OGG-08222 EXTRACT IE_CAPT must be registered with the database to perform this operation.

Try it:

SQL> select 'exec DBMS_CAPTURE_ADM.DROP_CAPTURE ('''||capture_name||''');' from dba_capture;

'EXECDBMS_CAPTURE_ADM.DROP_CAPTURE('''||CAPTURE_NAME||''');'
----------------------------------------------------------------------
exec DBMS_CAPTURE_ADM.DROP_CAPTURE ('OGG$CAP_IE_CAPT');

SQL> exec DBMS_CAPTURE_ADM.DROP_CAPTURE ('OGG$CAP_IE_CAPT');

PL/SQL procedure successfully completed.

Maiquel.

Database, RAC | Data Guard, Virtualization

VMWare: Adding Shared Disks for Clustered Oracle Database

 , , , ,

Hi folks!
Today a friend asked about how to configure disks on VMWare to create a virtualized cluster database. I revisited my old notes and decided to share. Here it goes…

First, I really have some constraints about it:
– Fake “high availability”: To have HA with VM it’s not needed 2 vms, if a host fail VMWare should make a VMotion (if well configured), and no services will be affected. So, one VM is ok.
– Not real “horizontally scallated”: It probably would be better to use one server as physical than have two vms on it. Not make sense to do it…

So, why?
To prove concept, evaluate RAC configuration (caches on sequences, etc) and labs, to learn and practice RAC stuffs…

Ok, now how to make it happen?

1. Add new disk to one of the machines. Some way, one will be the “primary” and share disks with another.
add1.jpg

2. Set Mode Thick Eager Zeroed

add2.jpg

3. Create a specific controller to this “shared disks”

add3.jpg

4. Set controller to virtual sharing

add4.jpg

# Other Machine
5. Adding the existent disk to other VM (not primary, but from primary)

add5.jpg
6. Select disk from primary

add6.jpg

add7.jpg

7. Create a new controller, as you made on primary and select it:

add8.jpgadd9.jpg

8. Set controller to virtual sharing
add10

OBS:
If this error happen, one of your controller is not in sharing mode. Please check it.
add11

 

And here we are! 🙂
Good lab!
Matheus.

Database, Linux, RAC | Data Guard

PRCR-1079 CRS-2674 CRS-5017 ORA-27102: out of memory Linux-x86_64 Error: 28: No space left on device

 , , ,

# Problem

myserver:/home/oracle>srvctl start database -d mydb
PRCR-1079 : Failed to start resource ora.mydb.db
CRS-5017: The resource action "ora.mydb.db start" encountered the following error:
ORA-27102: out of memory
Linux-x86_64 Error: 28: No space left on device
. For details refer to "(:CLSN00107:)" in "/grid/product/11.2.0/log/myserver/agent/crsd/oraagent_oracle/oraagent_oracle.log".
CRS-2674: Start of 'ora.mydb.db' on 'myserver' failed
CRS-5017: The resource action "ora.mydb.db start" encountered the following error:
ORA-27102: out of memory
Linux-x86_64 Error: 28: No space left on device
. For details refer to "(:CLSN00107:)" in "/grid/product/11.2.0/log/myserver2/agent/crsd/oraagent_oracle/oraagent_oracle.log".
CRS-2674: Start of 'ora.mydb.db' on 'myserver2' failed
CRS-2632: There are no more servers to try to place resource 'ora.mydb.db' on that would satisfy its placement policy
myserver:/home/oracle>

# Solution

On /etc/sysctl.conf ajust as below and then reload sysctl (“sysctl -p” as root):

#Old
#kernel.shmall = 24641536
#New
kernel.shmall = 4294967296

Matheus.

Database, Errors and Bugs, RAC | Data Guard

ORA-10456: cannot open standby database; media recovery session may be in progress

 , ,

A dataguard quick tip!

# Error

SQL> ALTER DATABASE OPEN READ ONLY;
ALTER DATABASE OPEN READ ONLY
*
ERROR at line 1:
ORA-10456: cannot open standby database; media recovery session may be in progress

# Solution

SQL> ALTER DATABASE RECOVER MANAGED STANDBY DATABASE CANCEL;
Database altered.
SQL> ALTER DATABASE OPEN READ ONLY;
Database altered.
SQL> ALTER DATABASE RECOVER MANAGED STANDBY DATABASE USING CURRENT LOGFILE DISCONNECT;
Database altered.

See ya! 🙂
Matheus.

Database, Errors and Bugs, Linux, RAC | Data Guard

VKTM Hang – High CPU Usage

 , , ,

Today a database (RHEL 6, single instance, 11.2.0.4) suddently started to “explode” CPU on VKTM process (100% CPU).
After some minutes lost (completely) in support.oracle.com (there was just a few notes about binary permissions on Solaris), I decided to make a McGayver by myself. 🙂

By Oracle words: “VKTM acts as a time publisher for an Oracle instance. VKTM publishes two sets of time: a wall clock time using a seconds interval and a higher resolution time (which is not wall clock time) for interval measurements. The VKTM timer service centralizes time tracking and offloads multiple timer calls from other clients.”

This way, my solution:

SQL> alter system set "_high_priority_processes"='LMS*' scope=spfile;
System altered.

And restart the database, of course.
So, VKTM is no more a “priority” process. The problem was solved. 🙂

Another possibility is to disable VKTM (undocumented parameter “_disable_vktm” – boolean). But I wanted to keep it running, changing less as possible of database configuration, just reducing priority.

KB:
Master Note: Troubleshooting Oracle Background Processes (Doc ID 1509616.1)
Great post about hidden parameters: http://oracleinaction.com/undocumented-params-11g/
Oficial one: http://www.orafaq.com/parms/index.htm

Hugs!
Matheus.

ASM, Database, Linux, RAC | Data Guard, Shell Scripts

Kludge: Mounting ACFS Thought Shellscript

 , , , , , ,

Just the script. The history is here.
This is a “workaround” script. As always, is recommended to use Oracle Restart, like I posted here.

#!/bin/sh
$GRID_HOME/bin/srvctl add filesystem -d /dev/asm/dbhome-270 -g 'DGHOME' -v DBHOME -m /oracle/db -u oracle
if [ $? = "0" -o $? = "2" ]; then
$GRID_HOME/bin/srvctl start filesystem -d /dev/asm/dbhome-270
if [ $? = "0" ]; then
chown oracle:oinstall /oracle/db
chmod 775 /oracle/db
$GRID_HOME/bin/srvctl status filesystem -d /dev/asm/dbhome-270
exit 0
fi
$GRID_HOME/bin/srvctl status filesystem -d /dev/asm/dbhome-270
fi

There is a good post ACFS and ACFS restart scripting:
https://levipereira.wordpress.com/2011/07/28/oracle-acfs-filesystem-managed-by-ohas-on-oracle-restart/

See ya!

Matheus.

Database, RAC | Data Guard, SQL | PLSQL

Scheduler Job by Node (RAC Database)

 , ,

Sometimes you want to run something just in one node of the RAC. Here is an example to do it:

create or replace procedure USER_JOB.PRC_SOMETHING is
begin
-- do something
null;
end;
/
begin
sys.dbms_scheduler.create_job(job_name => 'USER_JOB.JOB_SOMETHING',
job_type            => 'PLSQL_BLOCK',
job_action          => 'USER_JOB.PRC_SOMETHING;',
start_date          => sysdate,
repeat_interval     => 'Freq=Minutely;Interval=30',
end_date            => to_date(null),
job_class           => 'DEFAULT_JOB_CLASS',
enabled             => true,
auto_drop           => false,
comments            => 'Something Job.');
end;
/
begin
dbms_scheduler.set_attribute(name => 'USER_JOB.JOB_SOMETHING',
 attribute=>'INSTANCE_ID', value=> 1);
end;
/

Matheus.

ASM, Database, RAC | Data Guard, Tools

CRSCTL: AUTO_START of Cluster Services (ACFS)

 , , ,

As I sad long time ago (Manually Mounting ACFS)… Here is it:

To set autostart of a resource (in my case an ACFS) by CRSCTL, here the simple example:

# Check How it is currently configured:

[root@db1database1p bin]# ./crs_stat -p ora.dghome.dbhome.acfs |grep AUTO_START
AUTO_START=restore

# Set Autostart (and check):

[root@db1database1p bin]# ./crsctl modify resource ora.dghome.dbhome.acfs -attr AUTO_START=always
[root@db1database1p bin]# ./crs_stat -p ora.dghome.dbhome.acfs |grep AUTO_START
AUTO_START=always

It can be done also with “AUTO_START=1”. We have 3 possibilities (always, restore and never).

# KB
http://docs.oracle.com/cd/E11882_01/rac.112/e16794/resatt.htm#CWADD91444

Matheus.

Database, Errors and Bugs, RAC | Data Guard, SQL | PLSQL

ORA-02062: distributed recovery

 , , , ,

# Error/Alert

Errors in file /oracle/diag/rdbms/mydb/mydb2/trace/mydb2_reco_26083546.trc:
ORA-02062: distributed recovery received DBID e450df78, expected 0311e884

# Solution

begin
commit;
for d in (select local_tran_id from dba_2pc_pending) loop
dbms_transaction.purge_lost_db_entry( d.local_tran_id );
commit;
end loop;
end;
/

Matheus.