vaccum problem
От | Oleg Bartunov |
---|---|
Тема | vaccum problem |
Дата | |
Msg-id | Pine.GSO.3.96.SK.991011131943.4801V-100000@ra обсуждение исходный текст |
Список | pgsql-hackers |
Hi, I have a cron job to vacuum table which updates rather frequently and after month of work I'm getting NOTICE NOTICE: Index hits_pkey: NUMBER OF INDEX' TUPLES (10003) IS NOT THE SAME AS HEAP' (10006) I use psql -tq discovery < vacuum_hits.sql where vacuum_hits.sql is: vacuum analyze hits(msg_id); begin work; drop index hits_pkey; create unique index hits_pkey on hits(msg_id); end work; I rebuild index hits_pkey to avoid infinite grow - well, after Vadim's patch it still grows when table intensively updates. I've dumped and restored this table but I still get NOTICE message. The site I developed is in alpha stage, so sometimes I restart http server. I use persistent connection with postgres database so each http process has persistent connection with postgres database. Could be the problem If I just kill http processes and restart them ? Or I have to stop all postgres processes before ? What's the best way to manage persistent connections in 24*7 regime of http server ? Regards, Oleg PS. Forget to note: postgres 6.5.2, Linux 2.0.37, Apache 1.3.9, modperl 1.21, ApacheDBI 0.87 _____________________________________________________________ Oleg Bartunov, sci.researcher, hostmaster of AstroNet, Sternberg Astronomical Institute, Moscow University (Russia) Internet: oleg@sai.msu.su, http://www.sai.msu.su/~megera/ phone: +007(095)939-16-83, +007(095)939-23-83
В списке pgsql-hackers по дате отправления: