Android Development One More Time.

Hello, google this is C may I speak with Go please. Also, can you tell Cassandra there are a few tombstones laying around.

Good intentions going bad, Binding Views Unique indexing and noexpand

Yesterday was a very interesting day, trying to tune a procedure that uses a schema binding view, I try to add a needed unique clustered index into the view, so I can utilize the noexpand hint to used the actual indexes from the table, the result a total cluster ckfu. First there is not online support to build indexes on a view so the table got lock when adding the unique clustered index block the table, which cause everyone to jump in panic, while I was unaware of the situation as a let the script execute in a reduce script while working on another task. So lesson learns, monitor the execution of the script, if blocking way until a later time (maintenance window), the indexes will improved the execution to a 45% faster response time, but, careful planning should be in place, sometimes we get wrap into fixing and not realizing that everyone won't remember what you did good, but that one day when everything went to hell.

cheers.

PostgresSQL Scripts References.

A lot of work working on postgresql and today was another day of performance and fixing issues, eventually let add this set of scripts giving by one of my peers (Thanks Troy), now that I have them is time to shared with the world.

Cheers..

Is this a master or read only server I'm on now?  Can it accept writes?

-------------------------------------------------------------------------------------------------------

select pg_is_in_recovery();

or, just try to create a database.  You will fail if it's a standby.

-------------------------------------------------------------------------------------------------------

 

 

Check server activity:

-------------------------------------------------------------------------------------------------------

select * from pg_stat_activity;

-------------------------------------------------------------------------------------------------------

 

 

2 queries for database blocking and locking ( to be run on master node )

-------------------------------------------------------------------------------------------------------

SELECT blocked_locks.pid     AS blocked_pid,

blocked_activity.usename  AS blocked_user,

blocking_locks.pid     AS blocking_pid,

blocking_activity.usename AS blocking_user,

blocked_activity.query    AS blocked_statement,

blocking_activity.query   AS current_statement_in_blocking_process

FROM  pg_catalog.pg_locks         blocked_locks

JOIN pg_catalog.pg_stat_activity blocked_activity  ON blocked_activity.pid = blocked_locks.pid

JOIN pg_catalog.pg_locks         blocking_locks

ON blocking_locks.locktype = blocked_locks.locktype

AND blocking_locks.DATABASE IS NOT DISTINCT FROM blocked_locks.DATABASE

AND blocking_locks.relation IS NOT DISTINCT FROM blocked_locks.relation

AND blocking_locks.page IS NOT DISTINCT FROM blocked_locks.page

AND blocking_locks.tuple IS NOT DISTINCT FROM blocked_locks.tuple

AND blocking_locks.virtualxid IS NOT DISTINCT FROM blocked_locks.virtualxid

AND blocking_locks.transactionid IS NOT DISTINCT FROM blocked_locks.transactionid

AND blocking_locks.classid IS NOT DISTINCT FROM blocked_locks.classid

AND blocking_locks.objid IS NOT DISTINCT FROM blocked_locks.objid

AND blocking_locks.objsubid IS NOT DISTINCT FROM blocked_locks.objsubid

AND blocking_locks.pid != blocked_locks.pid

JOIN pg_catalog.pg_stat_activity blocking_activity ON blocking_activity.pid = blocking_locks.pid

WHERE NOT blocked_locks.GRANTED;

 

# There's also:

  

SELECT

waiting.locktype           AS waiting_locktype,

waiting.relation::regclass AS waiting_table,

waiting_stm.query          AS waiting_query,

waiting.mode               AS waiting_mode,

waiting.pid                AS waiting_pid,

other.locktype             AS other_locktype,

other.relation::regclass   AS other_table,

other_stm.query            AS other_query,

other.mode                 AS other_mode,

other.pid                  AS other_pid,

other.GRANTED              AS other_granted

FROM

pg_catalog.pg_locks AS waiting

JOIN

pg_catalog.pg_stat_activity AS waiting_stm

ON (waiting_stm.pid = waiting.pid)

JOIN

pg_catalog.pg_locks AS other

ON ((waiting."database" = other."database"

AND waiting.relation  = other.relation

)OR waiting.transactionid = other.transactionid)

JOIN pg_catalog.pg_stat_activity AS other_stm

ON (other_stm.pid = other.pid)

WHERE NOT waiting.GRANTED

AND waiting.pid <> other.pid;

-------------------------------------------------------------------------------------------------------

 

 

# Recursive view of blocking

-------------------------------------------------------------------------------------------------------

WITH RECURSIVE

     c(requested, CURRENT) AS

       ( VALUES

         ('AccessShareLock'::text, 'AccessExclusiveLock'::text),

         ('RowShareLock'::text, 'ExclusiveLock'::text),

         ('RowShareLock'::text, 'AccessExclusiveLock'::text),

         ('RowExclusiveLock'::text, 'ShareLock'::text),

         ('RowExclusiveLock'::text, 'ShareRowExclusiveLock'::text),

         ('RowExclusiveLock'::text, 'ExclusiveLock'::text),

         ('RowExclusiveLock'::text, 'AccessExclusiveLock'::text),

         ('ShareUpdateExclusiveLock'::text, 'ShareUpdateExclusiveLock'::text),

         ('ShareUpdateExclusiveLock'::text, 'ShareLock'::text),

         ('ShareUpdateExclusiveLock'::text, 'ShareRowExclusiveLock'::text),

         ('ShareUpdateExclusiveLock'::text, 'ExclusiveLock'::text),

         ('ShareUpdateExclusiveLock'::text, 'AccessExclusiveLock'::text),

         ('ShareLock'::text, 'RowExclusiveLock'::text),

         ('ShareLock'::text, 'ShareUpdateExclusiveLock'::text),

         ('ShareLock'::text, 'ShareRowExclusiveLock'::text),

         ('ShareLock'::text, 'ExclusiveLock'::text),

         ('ShareLock'::text, 'AccessExclusiveLock'::text),

         ('ShareRowExclusiveLock'::text, 'RowExclusiveLock'::text),

         ('ShareRowExclusiveLock'::text, 'ShareUpdateExclusiveLock'::text),

         ('ShareRowExclusiveLock'::text, 'ShareLock'::text),

         ('ShareRowExclusiveLock'::text, 'ShareRowExclusiveLock'::text),

         ('ShareRowExclusiveLock'::text, 'ExclusiveLock'::text),

         ('ShareRowExclusiveLock'::text, 'AccessExclusiveLock'::text),

         ('ExclusiveLock'::text, 'RowShareLock'::text),

         ('ExclusiveLock'::text, 'RowExclusiveLock'::text),

         ('ExclusiveLock'::text, 'ShareUpdateExclusiveLock'::text),

         ('ExclusiveLock'::text, 'ShareLock'::text),

         ('ExclusiveLock'::text, 'ShareRowExclusiveLock'::text),

         ('ExclusiveLock'::text, 'ExclusiveLock'::text),

         ('ExclusiveLock'::text, 'AccessExclusiveLock'::text),

         ('AccessExclusiveLock'::text, 'AccessShareLock'::text),

         ('AccessExclusiveLock'::text, 'RowShareLock'::text),

         ('AccessExclusiveLock'::text, 'RowExclusiveLock'::text),

         ('AccessExclusiveLock'::text, 'ShareUpdateExclusiveLock'::text),

         ('AccessExclusiveLock'::text, 'ShareLock'::text),

         ('AccessExclusiveLock'::text, 'ShareRowExclusiveLock'::text),

         ('AccessExclusiveLock'::text, 'ExclusiveLock'::text),

         ('AccessExclusiveLock'::text, 'AccessExclusiveLock'::text)

       ),

     l AS

       (

         SELECT

             (locktype,DATABASE,relation::regclass::text,page,tuple,virtualxid,transactionid,classid,objid,objsubid) AS target,

             virtualtransaction,

             pid,

             mode,

             GRANTED

           FROM pg_catalog.pg_locks

       ),

     t AS

       (

         SELECT

             blocker.target  AS blocker_target,

             blocker.pid     AS blocker_pid,

             blocker.mode    AS blocker_mode,

             blocked.target  AS target,

             blocked.pid     AS pid,

             blocked.mode    AS mode

           FROM l blocker

           JOIN l blocked

             ON ( NOT blocked.GRANTED

              AND blocker.GRANTED

              AND blocked.pid != blocker.pid

              AND blocked.target IS NOT DISTINCT FROM blocker.target)

           JOIN c ON (c.requested = blocked.mode AND c.CURRENT = blocker.mode)

       ),

     r AS

       (

         SELECT

             blocker_target,

             blocker_pid,

             blocker_mode,

             '1'::INT        AS depth,

             target,

             pid,

             mode,

             blocker_pid::text || ',' || pid::text AS seq

           FROM t

         UNION ALL

         SELECT

             blocker.blocker_target,

             blocker.blocker_pid,

             blocker.blocker_mode,

             blocker.depth + 1,

             blocked.target,

             blocked.pid,

             blocked.mode,

             blocker.seq || ',' || blocked.pid::text

           FROM r blocker

           JOIN t blocked

             ON (blocked.blocker_pid = blocker.pid)

           WHERE blocker.depth < 1000

       )

SELECT * FROM r

  ORDER BY seq;

-------------------------------------------------------------------------------------------------------

 

 

Need more info that the logs aren't giving you?  Adjust the postgresql.conf file to allow verbose logging.

You'll need to load the config changes by "pg_ctl reload", then check the logs to ensure they were loaded.

-------------------------------------------------------------------------------------------------------

log_duration = on|off

log_lock_waits = on|off

log_min_duration_statement = Xs (can be adjusted, in seconds)

deadlock_timeout = Xs (can be adjusted, in seconds)

-------------------------------------------------------------------------------------------------------

 

 

pg_buffercache (create extension pg_buffercache;) to create the extension if it isn't already there.

-------------------------------------------------------------------------------------------------------

SELECT c.relname, count(*) AS buffers

FROM pg_buffercache b INNER JOIN pg_class c

ON b.relfilenode = pg_relation_filenode(c.oid) AND

b.reldatabase IN (0, (SELECT oid FROM pg_database

WHERE datname = current_database()))

GROUP BY c.relname

ORDER BY 2 DESC

LIMIT 100;

-------------------------------------------------------------------------------------------------------

 

 

postgres cache hit ratio (create extension pg_buffercache;) to create the extension if it isn't already there.

-------------------------------------------------------------------------------------------------------

SELECT sum(heap_blks_read) as heap_read,

sum(heap_blks_hit)  as heap_hit,

sum(heap_blks_hit) / (sum(heap_blks_hit) + sum(heap_blks_read)) as ratio

FROM pg_statio_user_tables;

-------------------------------------------------------------------------------------------------------

 

 

postgres index cache hit ratio (create extension pg_buffercache;) to create the extension if it isn't already there.

-------------------------------------------------------------------------------------------------------

SELECT sum(idx_blks_read) as idx_read,

sum(idx_blks_hit)  as idx_hit,

(sum(idx_blks_hit) - sum(idx_blks_read)) / sum(idx_blks_hit) as ratio

FROM pg_statio_user_indexes;

-------------------------------------------------------------------------------------------------------

 

Is postgres using ssl for its connections?  Let's see:

-------------------------------------------------------------------------------------------------------

select datname, usename, client_addr, ssl, cipher

from pg_stat_activity

join pg_stat_ssl on pg_stat_activity.pid=pg_stat_ssl.pid;

-------------------------------------------------------------------------------------------------------

 

 

 

Table and index stats

-------------------------------------------------------------------------------------------------------

with table_stats as (

select psut.relname,

  psut.n_live_tup,

  1.0 * psut.idx_scan / greatest(1, psut.seq_scan + psut.idx_scan) as index_use_ratio

from pg_stat_user_tables psut

order by psut.n_live_tup desc

),

table_io as (

select psiut.relname,

  sum(psiut.heap_blks_read) as table_page_read,

  sum(psiut.heap_blks_hit)  as table_page_hit,

  sum(psiut.heap_blks_hit) / greatest(1, sum(psiut.heap_blks_hit) + sum(psiut.heap_blks_read)) as table_hit_ratio

from pg_statio_user_tables psiut

group by psiut.relname

order by table_page_read desc

),

index_io as (

select psiui.relname,

  psiui.indexrelname,

  sum(psiui.idx_blks_read) as idx_page_read,

  sum(psiui.idx_blks_hit) as idx_page_hit,

  1.0 * sum(psiui.idx_blks_hit) / greatest(1.0, sum(psiui.idx_blks_hit) + sum(psiui.idx_blks_read)) as idx_hit_ratio

from pg_statio_user_indexes psiui

group by psiui.relname, psiui.indexrelname

order by sum(psiui.idx_blks_read) desc

)

select ts.relname, ts.n_live_tup, ts.index_use_ratio,

  ti.table_page_read, ti.table_page_hit, ti.table_hit_ratio,

  ii.indexrelname, ii.idx_page_read, ii.idx_page_hit, ii.idx_hit_ratio

from table_stats ts

left outer join table_io ti

  on ti.relname = ts.relname

left outer join index_io ii

  on ii.relname = ts.relname

order by ti.table_page_read desc, ii.idx_page_read desc

;

-------------------------------------------------------------------------------------------------------

 

 

Table & Index Bloat query (from check_postgres script)

-------------------------------------------------------------------------------------------------------

SELECT

  current_database(), schemaname, tablename, /*reltuples::bigint, relpages::bigint, otta,*/

  ROUND((CASE WHEN otta=0 THEN 0.0 ELSE sml.relpages::FLOAT/otta END)::NUMERIC,1) AS tbloat,

  CASE WHEN relpages < otta THEN 0 ELSE bs*(sml.relpages-otta)::BIGINT END AS wastedbytes,

  iname, /*ituples::bigint, ipages::bigint, iotta,*/

  ROUND((CASE WHEN iotta=0 OR ipages=0 THEN 0.0 ELSE ipages::FLOAT/iotta END)::NUMERIC,1) AS ibloat,

  CASE WHEN ipages < iotta THEN 0 ELSE bs*(ipages-iotta) END AS wastedibytes

FROM (

  SELECT

    schemaname, tablename, cc.reltuples, cc.relpages, bs,

    CEIL((cc.reltuples*((datahdr+ma-

      (CASE WHEN datahdr%ma=0 THEN ma ELSE datahdr%ma END))+nullhdr2+4))/(bs-20::FLOAT)) AS otta,

    COALESCE(c2.relname,'?') AS iname, COALESCE(c2.reltuples,0) AS ituples, COALESCE(c2.relpages,0) AS ipages,

    COALESCE(CEIL((c2.reltuples*(datahdr-12))/(bs-20::FLOAT)),0) AS iotta -- very rough approximation, assumes all cols

  FROM (

    SELECT

      ma,bs,schemaname,tablename,

      (datawidth+(hdr+ma-(CASE WHEN hdr%ma=0 THEN ma ELSE hdr%ma END)))::NUMERIC AS datahdr,

      (maxfracsum*(nullhdr+ma-(CASE WHEN nullhdr%ma=0 THEN ma ELSE nullhdr%ma END))) AS nullhdr2

    FROM (

      SELECT

        schemaname, tablename, hdr, ma, bs,

        SUM((1-null_frac)*avg_width) AS datawidth,

        MAX(null_frac) AS maxfracsum,

        hdr+(

          SELECT 1+COUNT(*)/8

          FROM pg_stats s2

          WHERE null_frac<>0 AND s2.schemaname = s.schemaname AND s2.tablename = s.tablename

        ) AS nullhdr

      FROM pg_stats s, (

        SELECT

          (SELECT current_setting('block_size')::NUMERIC) AS bs,

          CASE WHEN SUBSTRING(v,12,3) IN ('8.0','8.1','8.2') THEN 27 ELSE 23 END AS hdr,

          CASE WHEN v ~ 'mingw32' THEN 8 ELSE 4 END AS ma

        FROM (SELECT version() AS v) AS foo

      ) AS constants

      GROUP BY 1,2,3,4,5

    ) AS foo

  ) AS rs

  JOIN pg_class cc ON cc.relname = rs.tablename

  JOIN pg_namespace nn ON cc.relnamespace = nn.oid AND nn.nspname = rs.schemaname AND nn.nspname <> 'information_schema'

  LEFT JOIN pg_index i ON indrelid = cc.oid

  LEFT JOIN pg_class c2 ON c2.oid = i.indexrelid

) AS sml

ORDER BY wastedbytes DESC;

-------------------------------------------------------------------------------------------------------

 

 

relations buffered in database share buffer

-------------------------------------------------------------------------------------------------------

select c.relname,pg_size_pretty(count(*) * 8192) as buffered,

        round(100.0 * count(*) / (

           select setting from pg_settings

           where name='shared_buffers')::integer,1)

        as buffer_percent,

        round(100.0*count(*)*8192 / pg_table_size(c.oid),1) as percent_of_relation

from pg_class c inner join pg_buffercache b on b.relfilenode = c.relfilenode inner

join pg_database d on ( b.reldatabase =d.oid and d.datname =current_database())

group by c.oid,c.relname order by 3 desc limit 100;

-------------------------------------------------------------------------------------------------------

 

 

disk usage

-------------------------------------------------------------------------------------------------------

select nspname,relname,pg_size_pretty(pg_relation_size(c.oid)) as "size"

from pg_class c left join pg_namespace n on ( n.oid=c.relnamespace)

where nspname not in ('pg_catalog','information_schema')

order by pg_relation_size(c.oid) desc limit 30;

-------------------------------------------------------------------------------------------------------

 

 

PostgreSql create weekly child partition table

It seems that doing a quarterly partition is causing issues with the auto-vacum and clean up process, so instead now we are creating a single child table each week and remove the data every seven days,
so it took no time to figure it out and created the function trigger, but there should be a better way somewhere, but for now we still using inherit tables instead of actual file partitions.
Cheers,


CREATE 
OR 
replace FUNCTION weeklypartition_insert() 
returns TRIGGER language 'plpgsql' cost 100 volatile NOT leakproof AS $body$
DECLARE _partition_year VARCHAR(25);_partition              varchar(80);_IsNewChild             boolean;_current_week           varchar(4);_tablename              varchar(80);_currentdate timestamp ;BEGIN
  _currentdate := CURRENT_TIMESTAMP; 
  -- get current week 
  _current_week := date_part('week',new.insertdate)::text; ----- Week of the year (1-52) 
  -- get YYYY_MM format 
  _partition_year := to_char (new.insertdate,'YYYY')||'_'||date_part('week',new.insertdate); 
  _partition := tg_table_name || '_' || _partition_year; 
  _tablename := _partition; 
  if NOT EXISTS 
  ( 
         SELECT 1 
         FROM   pg_tables 
         WHERE  schemaname= tg_table_schema 
         AND    tablename= _partition) THEN 
  raise notice 'A partition has been created %',tg_table_schema ||'.'|| _partition; 
  _isnewchild = true; 
  execute 'create table '|| tg_table_schema ||'.'|| _partition || ' (check( to_char (insertdate,''YYYY'||'_'|| _current_week||''') = '''|| to_char (new.insertdate,'YYYY')||'_'|| _current_week||''')) INHERITS ( ParentTableName);';
endIF;EXECUTE 'insert into '|| tg_table_schema || '.' || _partition || ' SELECT(' || tg_table_schema || '.' || tg_table_name || ' ' || quote_literal(new) || ').* RETURNING PrimaryKeyId;';IF (_isnewchild) then
-- Add primary key 
EXECUTE format 
  ('ALTER TABLE '||tg_table_schema||'.%s ADD PRIMARY KEY(PrimaryKeyId)', _partition); 
-- Add indexesEXECUTE format 
  ('CREATE INDEX ix_%s_PrimarykeyID ON '||tg_table_schema||'.%s USING btree ("SequenceID")', _tablename, _tablename);EXECUTE format
  ('CREATE INDEX ix_%s_Insertdate ON '||tg_table_schema||'.%s USING btree ("insertdate")', _tablename, _tablename);ENDIF;RETURN NULL;END;$BODY$;ALTER FUNCTION weeklypartition_insert() owner TO db_owner;
grant EXECUTE ON FUNCTION weeklypartition_insert() TO db_datawriter; 
grant EXECUTE ON FUNCTION weeklypartition_insert() TO PUBLIC; 
grant EXECUTE ON FUNCTION weeklypartition_insert() TO db_executor; 
grant EXECUTE ON FUNCTION weeklypartition_insert() TO db_owner;

Postgresql Insert into Multiple Tables using void.

When you think you know everything here comes something simple to keep your hands dirty, a developer try to translate a ms sql procedure into a postgresql function and eventually it was pass to me, so I wrote something similar to below with a special caveat.
Enjoy....

CREATE OR REPLACE FUNCTION SchemaName.FunctionName_Save(
                _parameter1 character varying DEFAULT NULL::character varying,
                _parameter2 character varying DEFAULT NULL::character varying,
                _parameter3 timestamp without time zone DEFAULT NULL::timestamp without time zone,
                __parameter4 timestamp without time zone DEFAULT NULL::timestamp without time zone,
                _parameter5 character varying DEFAULT NULL::character varying,
                _parameter6 character varying DEFAULT NULL::character varying,
                _parameter7 character varying DEFAULT NULL::character varying,
                _parameter8 character varying DEFAULT NULL::character varying,
                _parameter9 character varying DEFAULT NULL::character varying,
                _parameter0 character varying DEFAULT NULL::character varying,
                _time timestamp without time zone DEFAULT NULL::timestamp without time zone,
                _result character varying DEFAULT NULL::character varying,
                _details character varying DEFAULT NULL::character varying,
                _queue character varying DEFAULT NULL::character varying)
RETURNS void
    LANGUAGE 'plpgsql'
    COST 100
    VOLATILE

AS $BODY$

           declare
             _local01 int;
            _local02 int;
            _local03 int;
            _local04 int;
            _local05 bigint;
            _local06 bigint;
            _local07 bigint;
begin


    if not exists (  select 1 from SchemaName.TableName0 where lower (name) = lower(_parameter1) )
     THEN
               
                                insert into SchemaName.TableName0 (name,insertdate)
                                 values (_parameter1,__parameter4)
             returning SchemaName.TableName0.TableName0id into _local01;
      else
                 select SchemaName.TableName0.TableName0id into _local01 from  SchemaName.TableName0 where lower (ColumnName) = lower(_parameter1);
     end if;
    
    if not exists (  select 1 from  SchemaName.TableName01 where lower (ColumnName) = lower(_parameter2) )
     THEN
               
                                 insert into SchemaName.TableName01(name)
                                values (_parameter2)
             returning SchemaName.TableName01.TableName01id into _local02;
      else
                 select SchemaName.TableName01.TableName01id into _local02 from  SchemaName.TableName01 where lower (name) = lower(_parameter2);
     end if;
    
     if not exists (  select 1 from  SchemaName..TableName02 where lower (name) = lower(_parameter5) )
     THEN
               
                                 insert into SchemaName..TableName02(name)
                                values (_parameter5)
             returning SchemaName..TableName02..TableName02id into _local03;
      ELSE
        select SchemaName..TableName02..TableName02id into _local03 from  SchemaName..TableName02 where lower (ColumnName) = lower(_parameter5);
     end if;
    
     if not exists( select 1 from SchemaName.queue  where lower (queueaddress) = lower(_queue))
     then
        insert into SchemaName.queue(column,column)
       values (_queue,__parameter4)
       returning SchemaName.queue.queueid into _local04;
      else
        select SchemaName.queue.queueid into _local04 from SchemaName.queue  where lower (ColumnName) = lower(_queue);
     end if;
    
    _local05 := nextval('SchemaName.TableName04_seq'::regclass);
   
    INSERT INTO SchemaName.TableName04(
                   Columns,columns,columns)
       values
       ( _local02,_parameter6,_parameter3e,__parameter4,_local04);
      --- returning  SchemaName.TableName04.TableName04id into _local05;
      
       _local06 := nextval('SchemaName.TableName04_local06_seq'::regclass);
      INSERT INTO SchemaName.TableName04details(columns,columns,columns)
                    values( _local05,_parameter7,_parameter8,__parameter4)
       returning SchemaName.TableName04details.TableName04detailsid into _local06;
      
       INSERT INTO SchemaName..TableName06
    ( columns,columns,columns,columns,...)
    VALUES
    ( _local01, _parameter0, _parameter9, _local05, _Time, _local03, _Result, _Details);
  
end;

$BODY$;


ALTER FUNCTION SchemaName.FunctionName_Save(character varying, character varying, timestamp without time zone, timestamp without time zone, character varying, character varying, character varying, character varying, character varying, character varying, timestamp without time zone, character varying, character varying, character varying)
    OWNER TO db_owner;

GRANT EXECUTE ON FUNCTION SchemaName.FunctionName_Save(character varying, character varying, timestamp without time zone, timestamp without time zone, character varying, character varying, character varying, character varying, character varying, character varying, timestamp without time zone, character varying, character varying, character varying) TO db_datawriter;

GRANT EXECUTE ON FUNCTION SchemaName.FunctionName_Save(character varying, character varying, timestamp without time zone, timestamp without time zone, character varying, character varying, character varying, character varying, character varying, character varying, timestamp without time zone, character varying, character varying, character varying) TO PUBLIC;

GRANT EXECUTE ON FUNCTION SchemaName.FunctionName_Save(character varying, character varying, timestamp without time zone, timestamp without time zone, character varying, character varying, character varying, character varying, character varying, character varying, timestamp without time zone, character varying, character varying, character varying) TO db_executor;

GRANT EXECUTE ON FUNCTION SchemaName.FunctionName_Save(character varying, character varying, timestamp without time zone, timestamp without time zone, character varying, character varying, character varying, character varying, character varying, character varying, timestamp without time zone, character varying, character varying, character varying) TO db_owner;

SSIS Json Serializer for SQL Server 2014 and below

#region Help:  Introduction to the Script Component
/* The Script Component allows you to perform virtually any operation that can be accomplished in
 * a .Net application within the context of an Integration Services data flow.
 *
 * Expand the other regions which have "Help" prefixes for examples of specific ways to use
 * Integration Services features within this script component. */
#endregion

#region Namespaces
using System;
using System.Data;
using Microsoft.SqlServer.Dts.Pipeline.Wrapper;
using Microsoft.SqlServer.Dts.Runtime.Wrapper;
using Newtonsoft.Json;
using Newtonsoft.Json.Converters;
using System.IO;
using System.Text;
using System.Xml;
#endregion

/// <summary>
/// This is the class to which to add your code.  Do not change the name, attributes, or parent
/// of this class.
/// </summary>
[Microsoft.SqlServer.Dts.Pipeline.SSISScriptComponentEntryPointAttribute]
public class ScriptMain : UserComponent
{
    #region Help:  Using Integration Services variables and parameters
    /* To use a variable in this script, first ensure that the variable has been added to
     * either the list contained in the ReadOnlyVariables property or the list contained in
     * the ReadWriteVariables property of this script component, according to whether or not your
     * code needs to write into the variable.  To do so, save this script, close this instance of
     * Visual Studio, and update the ReadOnlyVariables and ReadWriteVariables properties in the
     * Script Transformation Editor window.
     * To use a parameter in this script, follow the same steps. Parameters are always read-only.
     *
     * Example of reading from a variable or parameter:
     *  DateTime startTime = Variables.MyStartTime;
     *
     * Example of writing to a variable:
     *  Variables.myStringVariable = "new value";
     */
    #endregion

    #region Help:  Using Integration Services Connnection Managers
    /* Some types of connection managers can be used in this script component.  See the help topic
     * "Working with Connection Managers Programatically" for details.
     *
     * To use a connection manager in this script, first ensure that the connection manager has
     * been added to either the list of connection managers on the Connection Managers page of the
     * script component editor.  To add the connection manager, save this script, close this instance of
     * Visual Studio, and add the Connection Manager to the list.
     *
     * If the component needs to hold a connection open while processing rows, override the
     * AcquireConnections and ReleaseConnections methods.
     * 
     * Example of using an ADO.Net connection manager to acquire a SqlConnection:
     *  object rawConnection = Connections.SalesDB.AcquireConnection(transaction);
     *  SqlConnection salesDBConn = (SqlConnection)rawConnection;
     *
     * Example of using a File connection manager to acquire a file path:
     *  object rawConnection = Connections.Prices_zip.AcquireConnection(transaction);
     *  string filePath = (string)rawConnection;
     *
     * Example of releasing a connection manager:
     *  Connections.SalesDB.ReleaseConnection(rawConnection);
     */
    #endregion

    #region Help:  Firing Integration Services Events
    /* This script component can fire events.
     *
     * Example of firing an error event:
     *  ComponentMetaData.FireError(10, "Process Values", "Bad value", "", 0, out cancel);
     *
     * Example of firing an information event:
     *  ComponentMetaData.FireInformation(10, "Process Values", "Processing has started", "", 0, fireAgain);
     *
     * Example of firing a warning event:
     *  ComponentMetaData.FireWarning(10, "Process Values", "No rows were received", "", 0);
     */
    #endregion

    /// <summary>
    /// This method is called once, before rows begin to be processed in the data flow.
    ///
    /// You can remove this method if you don't need to do anything here.
    /// </summary>
    public override void PreExecute()
    {
        base.PreExecute();
        /*
         * Add your code here
         */
    }

    /// <summary>
    /// This method is called after all the rows have passed through this component.
    ///
    /// You can delete this method if you don't need to do anything here.
    /// </summary>
    public override void PostExecute()
    {
        base.PostExecute();
        /*
         * Add your code here
         */
    }

    /// <summary>
    /// This method is called once for every row that passes through the component from Input0.
    ///
    /// Example of reading a value from a column in the the row:
    ///  string zipCode = Row.ZipCode
    ///
    /// Example of writing a value to a column in the row:
    ///  Row.ZipCode = zipCode
    /// </summary>
    /// <param name="Row">The row that is currently passing through the component</param>
    public override void Input0_ProcessInputRow(Input0Buffer Row)
    {
        try
        {
            
            
            var dlenght = Convert.ToInt32(Row.XmlData.Length);// Get Length side of xml record
         
            var ddata = Row.XmlData.GetBlobData(0, dlenght); // Get Data as a block of bytes 

            var xdata = Encoding.UTF8.GetString(ddata); // Get String Encoded to UFT8
            
            StringBuilder Rawjson = new StringBuilder(); // Get String Builder 


            XmlDataDocument doc = new XmlDataDocument(); // Create Xml document 

            doc.LoadXml(xdata); // Load Document 

            XmlElement root = doc.DocumentElement;

            /*
             * Block to add missong nodes to the  code or remove childrens
             * var xNode = doc.SelectSingleNode("Customer/Password");
            xNode.ParentNode.RemoveChild(xNode);
            var password = doc.CreateElement("Password");
            var email = doc.SelectSingleNode("Customer/Email");
            root.InsertBefore(password, email);
             * 
             */
                
            // this is where the magic begin
            var settings = new JsonSerializerSettings();
            settings.NullValueHandling = NullValueHandling.Include;
            
           

            byte[] jbyte = ASCIIEncoding.UTF8.GetBytes(JsonConvert.SerializeXmlNode(doc, Newtonsoft.Json.Formatting.None));
            Rawjson.Append(Row.CustomerId.ToString() + "|");
            Rawjson.Append(Encoding.UTF8.GetString(jbyte));
            jbyte = null;

            // json.Replace(Password, DBNull.Value.ToString());
            byte[] jsonbyte = ASCIIEncoding.UTF8.GetBytes(Rawjson.ToString());
            Output0Buffer.AddRow();
            Output0Buffer.CustomerId = (int)Row.CustomerId;
          
            Output0Buffer.Json.AddBlobData(jsonbyte);

        
        }
        catch (Exception ex)
        {
            String ErrorMesage = "Error_JsonConversation CustomerId:" + Row.CustomerId.ToString();

            bool fireError = true;

            IDTSComponentMetaData100 myMetaData;

            myMetaData = this.ComponentMetaData;

            myMetaData.FireError(0, ErrorMesage, ex.ToString(), string.Empty, 0, out fireError);
        }

    }

    public override void CreateNewOutputRows()
    {
        /*
          Add rows by calling the AddRow method on the member variable named "<Output Name>Buffer".
          For example, call MyOutputBuffer.AddRow() if your output was named "MyOutput".
        */
    }

}

PostgreSQL Quarterly Partition Function

Here is some quick code on how to create a partition function using inherit tables on PostgreSql, is simple but hopefully this will save some poor soul the documentation reading and for me save me some time in the future, currently things been very busy working in Ms Sql and PostgreSql doing several conversions from one DB to the other.
1a.-Create sequence
1b.- Create table
2.-Create function
3.-create trigger




-- FUNCTION:
 TableSchemaNameHere.TableName_partition_insert()

-- DROP FUNCTION TableSchemaNameHere.event_partiont_insert();
CREATE SEQUENCE seq_TableSchemaNameHere.TableNameHere_TableColumn
minvalue 1 nomaxvalue increment by 1;

create table TableSchemaNameHere.Tablename
(
 PKID int  DEFAULT nextval('seq_TableSchemaNameHere.TableNameHere_TableColumn') NOT NULL,
 insertdate timestamp(6) without time zone default now(),
 time timestamp without time zone,
 constraint pk_ primary key (pkid)
)

CREATE or Replace FUNCTION TableSchemaNameHere.TableName_partiont_insert()
    RETURNS trigger
    LANGUAGE 'plpgsql'
    COST 100.0
    VOLATILE NOT LEAKPROOF
   
AS $BODY$

                              declare
                                             declare
                                             _partition_date varchar(50);
               _partition varchar(80);
            _IsNewChild boolean;
             _current_day int;
            _first_day varchar(2) := '01';
            _last_day varchar(2);
            _pick_day varchar(2);
            _tablename varchar(80);

    begin
               -- get current day
        _current_day  := date_part('day',new.time);
     
        -- find last day of the month function
         _last_day := date_part ('day',TableSchemaNameHere.last_day (new.time::date));
        -- get YYYY_MM format
               _partition_date := to_char (NEW.time,'YYYY_MM');
     
        -- find if first or second sequence
        if _current_day < 16
         then
            _pick_day := _first_day;
            _partition_date  := _partition_date||'_'|| _first_day;
         
          else
           _pick_day = _last_day;
           _partition_date := _partition_date||'_'|| _last_day;
         end if;
       
       _partition := TG_TABLE_NAME || '_' || _partition_date;
       _tablename := _partition;
        if not exists ( select 1 from pg_tables where schemaname= TG_TABLE_SCHEMA and tablename= _partition)
        then
     
               RAISE NOTICE 'A partition has been created %',TG_TABLE_SCHEMA ||'.'|| _partition;
            _IsNewChild = TRUE;
            EXECUTE 'create table '|| TG_TABLE_SCHEMA ||'.'|| _partition || ' (check( to_char (time,''YYYY_MM'||'_'|| _pick_day||''') = '''|| to_char (NEW.time,'YYYY_MM')||'_'||_pick_day||''')) INHERITS ( TableSchemaNameHere.TableNamehere);';
         
        end if;
        execute 'insert into '|| TG_TABLE_SCHEMA || '.' || _partition || ' SELECT(' || TG_TABLE_SCHEMA || '.' || TG_TABLE_NAME || ' ' || quote_literal(NEW) || ').* RETURNING eventid;';
        if (_IsNewChild)
         then
           -- Add primary key
                                                            EXECUTE format('ALTER TABLE '||TG_TABLE_SCHEMA||'.%s ADD PRIMARY KEY(pkid)', _partition);
                -- Assign owner of of inherited table
                                                             EXECUTE format('ALTER TABLE '||TG_TABLE_SCHEMA||'.%s OWNER TO db_owner', _partition);

                                                         
                                                            -- Add FK promo_batch_id
                                                            EXECUTE format('ALTER TABLE '||TG_TABLE_SCHEMA||'.%s ADD CONSTRAINT FK_%s FOREIGN KEY(columnId) REFERENCES TableSchemaNameHere.fkid (fkid)  MATCH FULL',_tablename, _tablename);

                                                            -- Need to define indexes for inherited tables
                                                            EXECUTE format('CREATE INDEX ix_%s_EventTypeID ON '||TG_TABLE_SCHEMA||'.%s USING btree ("indexkeycolumn")', _tablename, _tablename);
             
                                                         
         end if;
        RETURN NULL;
    end
 
$BODY$;

----Create trigger
CREATE TRIGGER tr_TableName_Partition
BEFORE INSERT ON TableSchemaNameHere.TableNameHere
FOR EACH ROW EXECUTE PROCEDURE myschema.server_partition_function();