Change redo logs from filesystem to RAW

There is one thing I particularly like, but it is possible that we find people who say that if the database performance is greatly improved with raw redo in front of a filesystem (ext4 for example).

To find out for sure we can work with RAW mode redologs very easily. Let’s take an example made in Red Hat 4, with a test BDD with 3 redologs of 51 MB each:

Source   
[oracle@clu01 DBU]$ ls -l
total 1502076
-rw-r-----  1 oracle oinstall   9748480 Jun 25 20:38 control01.ctl
-rw-r-----  1 oracle oinstall   9748480 Jun 25 20:38 control02.ctl
-rw-r-----  1 oracle oinstall   7061504 Jun 12 11:36 control03.ctl
-rw-r-----  1 oracle oinstall  52429312 Jun 25 11:06 redo01.log
-rw-r-----  1 oracle oinstall  52429312 Jun 25 10:39 redo02.log
-rw-r-----  1 oracle oinstall  52429312 Jun 25 10:39 redo03.log
-rw-r-----  1 oracle oinstall 545267712 Jun 25 20:38 sysaux01.dbf
-rw-r-----  1 oracle oinstall 744497152 Jun 25 20:38 system01.dbf
-rw-r-----  1 oracle oinstall  20979712 Jun 23 21:22 temp01.dbf
-rw-r-----  1 oracle oinstall  36708352 Jun 25 20:38 undotbs01.dbf
-rw-r-----  1 oracle oinstall   5251072 Jun 25 20:38 users01.dbf

Continue reading

OpenLDAP checksum error and slapcat message “ldif_read_file: checksum error”

OpenLDAP has evolved allowing dynamic changes in the (very good in principle), which is usually stored in:

Source   
/etc/ldap/slapd.d

The recommendation is to use commands like: ldapadd, ldapmodify or ldapdelete.

But the reality is that often would be nice to make changes manually to end before, this is something you can do. Simply restarting the service then restart normally pretty fast.

The surprise comes when we do a:

Continue reading

LDAP query filtering by group

If we are integrating applications on an LDAP (in our case on a OpenLDAP), we have probably seen the utility to obtain the users belonging to one group even LDAP query, because we want to integrate the application supports only a single query.

Also if we documentation on this subject we have found stuff like, “for users belonging to the group grp_test run the query”.

Source   
memberOf=cn=grp_test,ou=Groups,dc=zentyal

Continue reading

MySQL jobs, scheduling tasks

This time we will see how to perform a scheduled task in MySQL (relatively speaking a job Oracle).

First we have to make sure we have the scheduler started with this we see:

Source   
mysql> SHOW processlist;
+----+------+-----------------+------------+---------+------+-------+------------------+
| Id | User | Host            | db         | Command | Time | State | Info             |
+----+------+-----------------+------------+---------+------+-------+------------------+
|  6 | root | localhost:49987 | assets_pru | Sleep   |  299 |       | NULL             |
|  8 | root | localhost       | assets_pru | Query   |    0 | NULL  | SHOW processlist |
+----+------+-----------------+------------+---------+------+-------+------------------+
2 rows IN SET (0.00 sec)

It is not started, for this we have to change a parameter of the mysqld section in my.cnf:

Continue reading

Apache SSL client certificate, configuring Apache to allow access only SSL certificate installed on client

Let’s configure Apache (on an Ubuntu 12) to allow access to clients with an installed SSL certificate personnel, first we have to create some structures to later work with revocation lists.

The first is to have openssl installed:

ubuntu@ip-10-112-31-82:~$ sudo aptitude install openssl

We will create a directory structure that conforms to the expected paths for the configuration file openssl.cnf:

Source   
ubuntu@ip-10-112-31-82:~$ mkdir -p /vol/apache2_certs
ubuntu@ip-10-112-31-82:~$ cd /vol/apache2_certs/
ubuntu@ip-10-112-31-82:/vol/apache2_certs$ sudo cp /etc/ssl/openssl.cnf .

The openssl.cnf file defines a directory structure to work among other things with lists of denial of certificates, will edit and modify the line:

Continue reading

rescan scsi linux

If we add disks in hot (from any system virtualization) it is possible that the OS does not know until we do a rescan of the SCSI bus, this can be done with the tool:

Source   
rescan-scsi-bus.sh -a

To install RedHat/Centos:

Source   
yum install sg3_utils

Continue reading

LVM external drive, HDD data access with LVM partition (eg a USB HD)

The LVM volumes have many advantages but when you connect a hard drive (with LVM) to an operating system (for example via USB) and want to access the data, we see that is not automatic.

To access the data directly we can mount the volume because the device simply does not exist, this can be easily solved. The sequence of actions is:

1- Connect the HD (logically)
2- Perform vgscan
3- Perform lvscan
4- Enable LVM volume desired
5- Mount the device and access the data

To extract HDD must:

Continue reading

MySQL tuning variables, parameters for InnoDB

Continuing the tuning MySQL will now play specifically InnoDB engine.

This post is linked to the “MySQL tuning parameters for any engine“, we will work from states and according a result see that variables can be modified to improve results.

In the post “MySQL variables states” already worked with variables and states.

Documentation of all the variables can be found in:
http://dev.mysql.com/doc/refman/5.5/en/dynamic-system-variables.html

Continue reading

Memory management for Oracle 10g and >= Oracle 11g

In 10g and 11g versions of Oracle Database, it has been simplified configuration memory structures for both the SGA and the PGA significantly.

Oracle version >=10g

From 10g, 2 new memory management parameters that enormously simplify this task are introduced.

SGA_TARGET, simply set a value and resizes demand values (if they are = 0):

  • Buffer cache (DB_CACHE_SIZE)
  • Shared pool (SHARED_POOL_SIZE)
  • Large pool (LARGE_POOL_SIZE)
  • Java pool (JAVA_POOL_SIZE)
  • Streams pool (STREAMS_POOL_SIZE)

Continue reading

MySQL tuning parameters for any engine

Many people make MySQL installations and do not care about the parameterization. The parameterization is important, one BD can work well (for now) with the default parameters, problems arise when the database grows or increases their workload.

In this post we will discuss the parameters that can affect the performance of any motor (the most used are InnoDB and MyISAM).

The variables are used by the server to size structures important memory for good engine performance of BD, states server will indicate if the variables we’ve defined are actually causing a positive effect or otherwise not suitable for absolutely nothing.

Continue reading