Differences

This shows you the differences between two versions of the page.

--- informatica:linux:zabbix [2014/07/29 06:29] – [Autenticacion LDAP] javi
+++ informatica:linux:zabbix [2015/03/14 12:47] – [Housekeeper y particionado] javi
@@ Line 116: / Line 116: @@
 . Crear el script de arranque:
-**TODO**: el script que se provee para debian creo recordar que no funciona del todo bien, en concreto la parte de "status".
   sudo ln -s /ruta/zabbix-2.0.2/misc/init.d/debian/zabbix-server /etc/init.d/
@@ Line 401: / Line 399: @@
 http://zabbixzone.com/zabbix/mysql-performance-tips-for-zabbix/
-Es MUY recomendable:
+http://dev.mysql.com/doc/refman/5.5/en/innodb-multiple-tablespaces.html
+https://www.zabbix.org/wiki/Docs/howto/mysql_partition
+. Habilitar file per table
+http://dev.mysql.com/doc/refman/5.5/en/innodb-multiple-tablespaces.html
+. Cambiar indices (puede ser lento y bloquear inserts):
+<code>
+mysql -u zabbix2014 -p
+use zabbix2014;
+Alter table history_text drop primary key, add index (id), drop index history_text_2, add index history_text_2 itemid, id);
+Alter table history_log drop primary key, add index (id), drop index history_log_2, add index history_log_2 (itemid, id);
+</code>
+. Crear procedures:
+<code>
+DELIMITER $$
+CREATE PROCEDURE `partition_create`(SCHEMANAME VARCHAR(64), TABLENAME VARCHAR(64), PARTITIONNAME VARCHAR(64), CLOCK INT)
+BEGIN
+        /*
+           SCHEMANAME = The DB schema in which to make changes
+           TABLENAME = The table with partitions to potentially delete
+           PARTITIONNAME = The name of the partition to create
+        */
+        /*
+           Verify that the partition does not already exist
+        */
+        DECLARE RETROWS INT;
+        SELECT COUNT(1) INTO RETROWS
+        FROM information_schema.partitions
+        WHERE table_schema = SCHEMANAME AND TABLE_NAME = TABLENAME AND partition_description >= CLOCK;
+        IF RETROWS = 0 THEN
+                /*
+. Print a message indicating that a partition was created.
+. Create the SQL to create the partition.
+. Execute the SQL from #2.
+                */
+                SELECT CONCAT( "partition_create(", SCHEMANAME, ",", TABLENAME, ",", PARTITIONNAME, ",", CLOCK, ")" ) AS msg;
+                SET @SQL = CONCAT( 'ALTER TABLE ', SCHEMANAME, '.', TABLENAME, ' ADD PARTITION (PARTITION ', PARTITIONNAME, ' VALUES LESS THAN (', CLOCK, '));' );
+                PREPARE STMT FROM @SQL;
+                EXECUTE STMT;
+                DEALLOCATE PREPARE STMT;
+        END IF;
+END$$
+DELIMITER ;
+</code>
+<code>
+DELIMITER $$
+CREATE PROCEDURE `partition_drop`(SCHEMANAME VARCHAR(64), TABLENAME VARCHAR(64), DELETE_BELOW_PARTITION_DATE BIGINT)
+BEGIN
+        /*
+           SCHEMANAME = The DB schema in which to make changes
+           TABLENAME = The table with partitions to potentially delete
+           DELETE_BELOW_PARTITION_DATE = Delete any partitions with names that are dates older than this one (yyyy-mm-dd)
+        */
+        DECLARE done INT DEFAULT FALSE;
+        DECLARE drop_part_name VARCHAR(16);
+        /*
+           Get a list of all the partitions that are older than the date
+           in DELETE_BELOW_PARTITION_DATE.  All partitions are prefixed with
+           a "p", so use SUBSTRING TO get rid of that character.
+        */
+        DECLARE myCursor CURSOR FOR
+                SELECT partition_name
+                FROM information_schema.partitions
+                WHERE table_schema = SCHEMANAME AND TABLE_NAME = TABLENAME AND CAST(SUBSTRING(partition_name FROM 2) AS UNSIGNED) < DELETE_BELOW_PARTITION_DATE;
+        DECLARE CONTINUE HANDLER FOR NOT FOUND SET done = TRUE;
+        /*
+           Create the basics for when we need to drop the partition.  Also, create
+           @drop_partitions to hold a comma-delimited list of all partitions that
+           should be deleted.
+        */
+        SET @alter_header = CONCAT("ALTER TABLE ", SCHEMANAME, ".", TABLENAME, " DROP PARTITION ");
+        SET @drop_partitions = "";
+        /*
+           Start looping through all the partitions that are too old.
+        */
+        OPEN myCursor;
+        read_loop: LOOP
+                FETCH myCursor INTO drop_part_name;
+                IF done THEN
+                        LEAVE read_loop;
+                END IF;
+                SET @drop_partitions = IF(@drop_partitions = "", drop_part_name, CONCAT(@drop_partitions, ",", drop_part_name));
+        END LOOP;
+        IF @drop_partitions != "" THEN
+                /*
+. Build the SQL to drop all the necessary partitions.
+. Run the SQL to drop the partitions.
+. Print out the table partitions that were deleted.
+                */
+                SET @full_sql = CONCAT(@alter_header, @drop_partitions, ";");
+                PREPARE STMT FROM @full_sql;
+                EXECUTE STMT;
+                DEALLOCATE PREPARE STMT;
+                SELECT CONCAT(SCHEMANAME, ".", TABLENAME) AS `table`, @drop_partitions AS `partitions_deleted`;
+        ELSE
+                /*
+                   No partitions are being deleted, so print out "N/A" (Not applicable) to indicate
+                   that no changes were made.
+                */
+                SELECT CONCAT(SCHEMANAME, ".", TABLENAME) AS `table`, "N/A" AS `partitions_deleted`;
+        END IF;
+END$$
+DELIMITER ;
+</code>
+<code>
+DELIMITER $$
+CREATE PROCEDURE `partition_maintenance`(SCHEMA_NAME VARCHAR(32), TABLE_NAME VARCHAR(32), KEEP_DATA_DAYS INT, HOURLY_INTERVAL INT, CREATE_NEXT_INTERVALS INT)
+BEGIN
+        DECLARE OLDER_THAN_PARTITION_DATE VARCHAR(16);
+        DECLARE PARTITION_NAME VARCHAR(16);
+        DECLARE LESS_THAN_TIMESTAMP INT;
+        DECLARE CUR_TIME INT;
+        CALL partition_verify(SCHEMA_NAME, TABLE_NAME, HOURLY_INTERVAL);
+        SET CUR_TIME = UNIX_TIMESTAMP(DATE_FORMAT(NOW(), '%Y-%m-%d 00:00:00'));
+        SET @__interval = 1;
+        create_loop: LOOP
+                IF @__interval > CREATE_NEXT_INTERVALS THEN
+                        LEAVE create_loop;
+                END IF;
+                SET LESS_THAN_TIMESTAMP = CUR_TIME + (HOURLY_INTERVAL * @__interval * 3600);
+                SET PARTITION_NAME = FROM_UNIXTIME(CUR_TIME + HOURLY_INTERVAL * (@__interval - 1) * 3600, 'p%Y%m%d%H00');
+                CALL partition_create(SCHEMA_NAME, TABLE_NAME, PARTITION_NAME, LESS_THAN_TIMESTAMP);
+                SET @__interval=@__interval+1;
+        END LOOP;
+        SET OLDER_THAN_PARTITION_DATE=DATE_FORMAT(DATE_SUB(NOW(), INTERVAL KEEP_DATA_DAYS DAY), '%Y%m%d0000');
+        CALL partition_drop(SCHEMA_NAME, TABLE_NAME, OLDER_THAN_PARTITION_DATE);
+END$$
+DELIMITER ;
+</code>
+<code>
+DELIMITER $$
+CREATE PROCEDURE `partition_verify`(SCHEMANAME VARCHAR(64), TABLENAME VARCHAR(64), HOURLYINTERVAL INT(11))
+BEGIN
+        DECLARE PARTITION_NAME VARCHAR(16);
+        DECLARE RETROWS INT(11);
+        DECLARE FUTURE_TIMESTAMP TIMESTAMP;
+        /*
+         * Check if any partitions exist for the given SCHEMANAME.TABLENAME.
+         */
+        SELECT COUNT(1) INTO RETROWS
+        FROM information_schema.partitions
+        WHERE table_schema = SCHEMANAME AND TABLE_NAME = TABLENAME AND partition_name IS NULL;
+        /*
+         * If partitions do not exist, go ahead and partition the table
+         */
+        IF RETROWS = 1 THEN
+                /*
+                 * Take the current date at 00:00:00 and add HOURLYINTERVAL to it.  This is the timestamp below which we will store values.
+                 * We begin partitioning based on the beginning of a day.  This is because we don't want to generate a random partition
+                 * that won't necessarily fall in line with the desired partition naming (ie: if the hour interval is 24 hours, we could
+                 * end up creating a partition now named "p201403270600" when all other partitions will be like "p201403280000").
+                 */
+                SET FUTURE_TIMESTAMP = TIMESTAMPADD(HOUR, HOURLYINTERVAL, CONCAT(CURDATE(), " ", '00:00:00'));
+                SET PARTITION_NAME = DATE_FORMAT(CURDATE(), 'p%Y%m%d%H00');
+                -- Create the partitioning query
+                SET @__PARTITION_SQL = CONCAT("ALTER TABLE ", SCHEMANAME, ".", TABLENAME, " PARTITION BY RANGE(`clock`)");
+                SET @__PARTITION_SQL = CONCAT(@__PARTITION_SQL, "(PARTITION ", PARTITION_NAME, " VALUES LESS THAN (", UNIX_TIMESTAMP(FUTURE_TIMESTAMP), "));");
+                -- Run the partitioning query
+                PREPARE STMT FROM @__PARTITION_SQL;
+                EXECUTE STMT;
+                DEALLOCATE PREPARE STMT;
+        END IF;
+END$$
+DELIMITER ;
+</code>
+Este ultimo los gobierna todos, por tanto hay que seleccionar los valores que se adapte a nuestras necesidades:
+<code>
+DELIMITER $$
+CREATE PROCEDURE `partition_maintenance_all`(SCHEMA_NAME VARCHAR(32))
+BEGIN
+                CALL partition_maintenance(SCHEMA_NAME, 'history', 28, 24, 14);
+                CALL partition_maintenance(SCHEMA_NAME, 'history_log', 28, 24, 14);
+                CALL partition_maintenance(SCHEMA_NAME, 'history_str', 28, 24, 14);
+                CALL partition_maintenance(SCHEMA_NAME, 'history_text', 28, 24, 14);
+                CALL partition_maintenance(SCHEMA_NAME, 'history_uint', 28, 24, 14);
+                CALL partition_maintenance(SCHEMA_NAME, 'trends', 730, 24, 14);
+                CALL partition_maintenance(SCHEMA_NAME, 'trends_uint', 730, 24, 14);
+END$$
+DELIMITER ;
+</code>
+. Dar permisos de ejecucion al ultimo procedimiento. Vamos a dar al mismo usuario que puede leer la base de datos de zabbix:
+<code>
+GRANT EXECUTE ON PROCEDURE zabbix2014.partition_maintenance_all TO 'zabbix2014'@'localhost';
+flush privileges;
+exit;
+</code>
+. Creamos archivo para poder iniciar sesion mysql sin contrasenya:
+  vim ~/.my.cnf
+Con el siguiente contenido:
+<code>
+[client]
+user=zabbix2014
+password=zabbix2014
+</code>
+Ajustamos permisos:
+  chmod 0600 ~/.my.cnf
+. Creamos un script para ser llamado desde cron:
+<code>
+sudo vim /opt/partition_maintenance_all.sh
+</code>
+Con el siguiente contenido (ajustar variables):
+<code>
+#!/bin/bash
+LOG="/var/log/"`basename $0`".log"
+STRING_START="START of the script:"
+STRING_END="END of the script:"
+MYSQL_USER="zabbix2014"
+MYSQL_DB=$MYSQL_USER
+MYSQL_PROCEDURE="partition_maintenance_all"
+[[ ! $LOG ]]; touch $LOG
+DATE="$(date +%d/%m/%Y) $(date +%H:%M:%S)"
+echo -e "\n################### $STRING_START $DATE ##################"  >> $LOG
+#mysql -u $MYSQL_USER -e "CALL partition_maintenance_all('zabbix2014')\G" zabbix2014 >> $LOG 2>&1
+#mysql -u $MYSQL_USER -e "CALL partition_maintenance_all('"$MYSQL_DB"')\G" $MYSQL_DB >> $LOG 2>&1
+mysql -u $MYSQL_USER -e "CALL "$MYSQL_PROCEDURE"('"$MYSQL_DB"')\G" $MYSQL_DB >> $LOG 2>&1
+DATE="$(date +%d/%m/%Y) $(date +%H:%M:%S)"
+echo -e "\n################### $STRING_END $DATE ##################"  >> $LOG
+</code>
+Ajustamos permisos y logs:
+<code>
+sudo chown usuario:usuario /opt/partition_maintenance_all.sh
+chmod +x /opt/partition_maintenance_all.sh
+sudo touch /var/log/partition_maintenance_all.sh.log
+sudo chown root:usuario /var/log/partition_maintenance_all.sh.log
+chmod 660 /var/log/partition_maintenance_all.sh.log
+</code>
-. [[http://dev.mysql.com/doc/refman/5.5/en/innodb-multiple-tablespaces.html|InnoDB File-Per-Table Mode]]
+Y anyadimos la entrada a cron, para ser ejecutada, por ejemplo, cada dia:
-. [[https://www.zabbix.org/wiki/Docs/howto/mysql_partition|mysql partition]]
+4 * * * /opt/partition_maintenance_all.sh >> /var/log/partition_maintenance_all.sh.log 2>&1
 ====== Comandos varios ======