> Bom dia,
>
>
> Atualizei um servidor que estava utilizando a versão 9.0 para a 9.4.7
> e após atualização esta query passou a ficar extremamente lenta.
>
> SQL:
> select movto_lote.nr_dcto
> from fin_receber_parc parc
> inner join fin_receber receber on (receber.cd_movto = parc.cd_movto_rec and
> receber.cd_serie_lcto =
> parc.cd_serie_lcto)
> inner join fin_receber_cnt cnt on (cnt.cd_movto_rec= receber.cd_movto and
>
> cnt.cd_serie_lcto=receber.cd_serie_lcto)
> inner join cnt_movto movto on (cnt.cd_movto=movto.cd_movto and
> movto.cd_serie_lcto=cnt.cd_serie_lcto)
> inner join cnt_movto_lote movto_lote on (movto_lote.cd_movto =
> movto.cd_movto_lote and
> movto_lote.cd_serie_lcto =
> movto.cd_serie_lcto)
> inner join cnt_competencia competencia on (competencia.cd_competencia
> =
> movto_lote.cd_competencia)
> inner join uni_emitente emitente on (emitente.cd_emitente =
> receber.cd_devedor)
> inner join fin_especie especie on (especie.cd_especie =
> parc.cd_especie ) where upper(coalesce(movto_lote.ds_historico, '%')) like
> upper('%%') and
> upper(coalesce(movto_lote.nr_dcto, '%')) like upper('%%') and
> upper(coalesce(competencia.nm_competencia, '%')) like upper('%%') and
> upper(coalesce(emitente.nm_emitentecompleto, '%')) like
> upper('%consumidor%') and
> coalesce(parc.vl_valor, 0) between coalesce(0, 0) and
> coalesce(999999.99, 999999999) and
> coalesce(receber.cd_devedor, 0) between coalesce(0, 0) and
> coalesce(999999999, 999999999) and
> movto_lote.dt_movto between ('01/01/2000') and ('01/03/9999') and
> parc.dt_vcto between ('01/01/2000') and ('31/12/9999') and
> parc.st_quitado = false and
> movto_lote.cd_estabelecimento=1
> order by parc.dt_vcto desc
>
>
> Saída do Explain Analyze:
> "Sort (cost=4779.91..4779.91 rows=1 width=16) (actual
> time=128286.342..128286.342 rows=6 loops=1)"
> " Sort Key: parc.dt_vcto DESC"
> " Sort Method: quicksort Memory: 25kB"
> " -> Nested Loop (cost=1.96..4779.90 rows=1 width=16) (actual
> time=126064.707..128286.279 rows=6 loops=1)"
> " Join Filter: (parc.cd_especie = especie.cd_especie)"
> " Rows Removed by Join Filter: 36"
> " -> Nested Loop (cost=1.96..4778.74 rows=1 width=18) (actual
> time=126064.672..128286.198 rows=6 loops=1)"
> " -> Nested Loop (cost=1.68..4777.55 rows=1 width=22) (actual
> time=64031.477..128247.981 rows=3189 loops=1)"
> " Join Filter: (movto_lote.cd_competencia =
> competencia.cd_competencia)"
> " Rows Removed by Join Filter: 15945"
> " -> Nested Loop (cost=1.68..4776.37 rows=1 width=26)
> (actual time=64031.432..128164.285 rows=3189 loops=1)"
> " Join Filter: (receber.cd_serie_lcto =
> movto_lote.cd_serie_lcto)"
> " -> Nested Loop (cost=1.26..4775.86 rows=1
> width=26) (actual time=0.085..128005.877 rows=6277 loops=1)"
> " Join Filter: (receber.cd_serie_lcto =
> movto.cd_serie_lcto)"
> " -> Nested Loop (cost=0.83..4774.95 rows=1
> width=24) (actual time=0.078..127775.182 rows=6277 loops=1)"
> " -> Nested Loop (cost=0.42..3012.16
> rows=1 width=30) (actual time=0.067..214.384 rows=6318 loops=1)"
> " -> Seq Scan on
> fin_receber_parc parc (cost=0.00..2745.68 rows=32 width=16) (actual
> time=0.052..91.802 rows=6318 loops=1)"
> " Filter: ((NOT st_quitado)
> AND (COALESCE(vl_valor, '0'::numeric) >= '0'::numeric) AND
> (COALESCE(vl_valor, '0'::numeric) <= 999999.99) AND (dt_vcto >=
> '2000-01-01'::date) AND (dt_vcto <= '9999-12-31'::date))"
> " Rows Removed by Filter:
> 81216"
> " -> Index Scan using
> "pk-fin_receber-geral" on fin_receber receber (cost=0.42..8.32 rows=1
> width=14) (actual time=0.013..0.014 rows=1 loops=6318)"
> " Index Cond: ((cd_movto =
> parc.cd_movto_rec) AND (cd_serie_lcto = parc.cd_serie_lcto))"
> " Filter:
> ((COALESCE(cd_devedor, 0) >= 0) AND (COALESCE(cd_devedor, 0) <= 999999999))"
> " -> Index Scan using
> "pk-fin_receber_cnt-cd_movto-cd_serie_lcto" on fin_receber_cnt cnt
> (cost=0.42..1762.78 rows=1 width=18) (actual time=7.918..20.187 rows=1
> loops=6318)"
> " Index Cond: (cd_serie_lcto =
> receber.cd_serie_lcto)"
> " Filter: (receber.cd_movto =
> cd_movto_rec)"
> " Rows Removed by Filter: 87707"
> " -> Index Scan using
> "pk-cnt_movto-cd_movto-cd_serie_lcto" on cnt_movto movto
> (cost=0.42..0.90
> rows=1 width=18) (actual time=0.031..0.031 rows=1 loops=6277)"
> " Index Cond: ((cd_movto =
> cnt.cd_movto) AND (cd_serie_lcto = cnt.cd_serie_lcto))"
> " -> Index Scan using "pk-cnt_movto_lote-geral" on
> cnt_movto_lote movto_lote (cost=0.42..0.50 rows=1 width=26) (actual
> time=0.022..0.023 rows=1 loops=6277)"
> " Index Cond: ((cd_movto =
> movto.cd_movto_lote) AND (cd_serie_lcto = movto.cd_serie_lcto))"
> " Filter: ((dt_movto >= '2000-01-01
> 00:00:00-02'::timestamp with time zone) AND (dt_movto <= '9999-03-01
> 00:00:00-03'::timestamp with time zone) AND (cd_estabelecimento = 1)
> AND (upper((COALESCE(ds_historico, '%'::character va (...)"
> " Rows Removed by Filter: 0"
> " -> Seq Scan on cnt_competencia competencia
> (cost=0.00..1.11 rows=6 width=4) (actual time=0.006..0.012 rows=6
> loops=3189)"
> " Filter: (upper((COALESCE(nm_competencia,
> '%'::bpchar))::text) ~~ '%%'::text)"
> " -> Index Scan using "pk-uni_emitente-cd_emitente" on
> uni_emitente emitente (cost=0.28..1.18 rows=1 width=4) (actual
> time=0.010..0.010 rows=0 loops=3189)"
> " Index Cond: (cd_emitente = receber.cd_devedor)"
> " Filter: (upper((COALESCE(nm_emitentecompleto,
> '%'::character varying))::text) ~~ '%CONSUMIDOR%'::text)"
> " Rows Removed by Filter: 1"
> " -> Seq Scan on fin_especie especie (cost=0.00..1.07 rows=7
> width=4) (actual time=0.005..0.006 rows=7 loops=6)"
> "Planning time: 22.367 ms"
> "Execution time: 128286.670 ms"
>
>
> Pelo que estou entendendo o problema está no fin_receber_cnt. Mas não
> to achando o furo.
> Observem que na versão 9.0 estava funcionando de forma satisfatoria.
>
Atualize as estatísticas com o comando ANALYZE.
http://www.postgresql.org/docs/9.4/interactive/sql-analyze.html
Por favor, me explique como vc chegou a esta conclusão? diz algo aí que as
estatísticas estão desatualizadas? (eu não manjo muito da saída do analyze e to
me batendo nisso a horas).
Mas eu havia feito todas as manutenções antes de enviar a saída do analyze e
inclusive peguei o banco e subi numa máquina minha e continua na mesma lentidão.
Talvez precise que eu envie mais alguma informação pra ajudar?
_______________________________________________
pgbr-geral mailing list
pgbr-geral@listas.postgresql.org.br
https://listas.postgresql.org.br/cgi-bin/mailman/listinfo/pgbr-geral
