Olá Pessoal,
Estou com uma dúvida, por opção nossa resolvemos não deixar o autovacuum no
automático e sim no manual (cron nas madrugadas), porém percebemos que ainda
assim vemos com frequencia vários processos com o autovacuum, será que
fizemos algo em desacordo?
Estamos enfrentando uma dificuldade (antes não tinha) de quando estes
processos estarem rodando as tabelas envolvidas estarem locadas e assim
fica, até que um de nós mata o pid da operação (autovacuum).
NOSSO CENÁRIO
PostgreSQL 8.3.7 on x86_64-pc-linux-gnu, compiled by GCC gcc-4.3.real
(Ubuntu 4.3.2-1ubuntu11) 4.3.2
PS AUX|GREP POSTGRES
postgres 5262 0.0 0.0 72168 1368 ? Ss Aug16 1:08 postgres:
logger process
root 25377 0.0 0.0 70124 3136 ? Ss 07:09 0:00 sshd:
postgres [priv]
postgres 25379 0.0 0.0 70124 1896 ? S 07:09 0:00 sshd:
postg...@pts/0
postgres 25380 0.0 0.0 19480 2272 pts/0 Ss 07:09 0:00 -bash
root 25385 0.0 0.0 70124 3140 ? Ss 07:09 0:00 sshd:
postgres [priv]
postgres 25387 0.0 0.0 70124 1900 ? S 07:09 0:00 sshd:
postg...@pts/1
postgres 25388 0.0 0.0 19480 2288 pts/1 Ss 07:09 0:00 -bash
postgres 25407 0.0 0.0 21016 1804 pts/0 S+ 07:10 0:00 vim
teste.py
postgres 25587 0.0 0.1 703176 7752 ? Ss 07:25 0:00 postgres:
writer process
postgres 25588 0.0 0.0 703176 1568 ? Ss 07:25 0:00 postgres:
wal writer process
postgres 25589 0.0 0.0 703176 1712 ? Ss 07:25 0:00 postgres:
autovacuum launcher process
postgres 25590 0.0 0.0 72296 1300 ? Ss 07:25 0:00 postgres:
archiver process
postgres 25591 0.0 0.0 72296 1432 ? Ss 07:25 0:00 postgres:
stats collector process
postgres 25595 0.1 0.3 710820 15052 ? Ss 07:27 0:02 postgres:
postgres apoiorh 10.4.1.42(3856) idle in transaction
postgres 25684 0.0 0.0 16152 1144 pts/1 R+ 07:54 0:00 ps aux
postgres 25685 0.0 0.0 7500 816 pts/1 R+ 07:54 0:00 grep
postgres
POSTGRESQL.CONF
#---------------------------------------------------------------------------
---
383 # AUTOVACUUM PARAMETERS
384
#---------------------------------------------------------------------------
---
385
386 #autovacuum = off # Enable autovacuum subprocess? 'on'
387 # requires track_counts to also be on.
388 #log_autovacuum_min_duration = -1 # -1 disables, 0 logs all actions
and
389 # their durations, > 0 logs only
390 # actions running at least that time.
391 #autovacuum_max_workers = 3 # max number of autovacuum
subprocesses
392 #autovacuum_naptime = 1min # time between autovacuum runs
393 #autovacuum_vacuum_threshold = 500 # min number of row updates
before
394 # vacuum
395 #autovacuum_analyze_threshold = 500 # min number of row updates
before
396 # analyze
397 #autovacuum_vacuum_scale_factor = 0.2 # fraction of table size
before vacuum
398 #autovacuum_analyze_scale_factor = 0.1 # fraction of table size
before analyze
399 #autovacuum_freeze_max_age = 200000000 # maximum XID age before
forced vacuum
400 # (change requires restart)
401 #autovacuum_vacuum_cost_delay = 20 # default vacuum cost delay for
402 # autovacuum, -1 means use
403 # vacuum_cost_delay
404 #autovacuum_vacuum_cost_limit = -1 # default vacuum cost limit for
405 # autovacuum, -1 means use
406 # vacuum_cost_limit
407
408
409
#---------------------------------------------------------------------------
---
Att.
----------------------------------------------------------------------------
Rubens José Rodrigues
" Falta de tempo é desculpa daqueles que perdem tempo por falta de métodos"
_______________________________________________
pgbr-geral mailing list
[email protected]
https://listas.postgresql.org.br/cgi-bin/mailman/listinfo/pgbr-geral
