| Type: | integer |
| Défaut: | 20 (20ms) |
| Min: | -1 (-1) |
| Max: | 100 (100ms) |
| Unité: | milliseconds (ms) |
| Contexte: | sighup |
| Redémarrer: | false |
Indique la valeur du coût de délai utilisée dans les opérations de VACUUM. Si -1 est indiqué, la valeur habituelle de vacuum_cost_delay est utilisée. La valeur par défaut est 20 millisecondes. Ce paramètre ne peut être configuré que dans le fichier postgresql.conf ou indiqué sur la ligne de commande mais il est possible de le surcharger pour toute table en modifiant les paramètres de stockage de la table.
Recommandations [EN]
If autovacuum is having too much of a performance impact on running queries, you might want to increase this setting to 50ms. However, this will also cause individual vacuum tasks to take longer.
Sur StackOverflow
- What is autovacuum_vacuum_cost_delay in autovacuum in PostgreSQL?
- Postgres auto-vacuum doesn't reclaim the dead tuples space causes disk full issue
- is there any adverse effect on DB if I set autovacuum scale factor to zero for certain tables?
- Need clarification related to Autovacuum configuration parameters
- vacuum_cost_page_miss set to zero
Sur pgsql-hackers
- Running autovacuum dynamic update to cost_limit and delay
- Re: [BUG] Autovacuum not dynamically decreasing cost_limit and cost_delay
- We should stop telling users to "vacuum that database in single-user mode"
- automatic analyze: readahead - add "IO read time" log message
- Re: automatic analyze: readahead - add "IO read time" log message