Deadlock detected during vacuum analyze - Mailing list pgsql-general

From Együd Csaba
Subject Deadlock detected during vacuum analyze
Date
Msg-id 0I8Q00LVF8IQFF@mail.vnet.hu
Whole thread Raw
List pgsql-general
Hi,
one more question: how can I set up vaccum handling in postgres to make it
less agressive? I'm faceing a problem where a batch process (inserting 184
records at once which takes 1 second in normal cases) stops because of a
dead lock. This occures (seemingly) only when a vacuum analyze is running.
Stopping vacuum calls, the problem thisappeares.

I set the vacuum_cost_delay option higher, but it doesn't help. I thought
that the more higher delay between vacuum portions the more time to finish
the inserts. Other settings have the default values.

To tell the truth I do not really understand this "cost" thing. Should I
treat it as a percent-like value?


My vacuum settings:

vacuum_cost_delay = 30
#vacuum_cost_page_hit = 1       # 0-10000 credits
#vacuum_cost_page_miss = 10    # 0-10000 credits
#vacuum_cost_page_dirty = 20    # 0-10000 credits
#vacuum_cost_limit = 200    # 0-10000 credits

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.805 / Virus Database: 547 - Release Date: 2004.12.03.



pgsql-general by date:

Previous
From: Michael Fuhr
Date:
Subject: Re: Copy row from table to table
Next
From: "Dann Corbit"
Date:
Subject: Re: Corrupt RTREE index