Re: [GENERAL] search_path reset on error (making it not reset)

2012-07-21 Thread Craig Ringer
On 07/21/2012 12:42 PM, imagene...@gmail.com wrote: The search_path set for the connection is being reset if a query errors. Is there a setting to maintain the search_path despite errors. Are you setting `search_path` within a transaction? Set it outside an explicit transaction if you want it

[GENERAL] Why is an ISO-8859-8 database allowing values not within that set?

2012-07-21 Thread Herouth Maoz
I am using Postgresql 8.3.14 on our reporting system. There are scripts that collect data from many databases across the firm into this database. Recently I added tables from a particular database which has encoding UTF-8. My dump procedure says \encoding ISO-8859-8 \copy ( SELECT ... ) to file

Re: [GENERAL] High CPU on read-only streaming replication server

2012-07-21 Thread Craig Ringer
On 07/20/2012 04:44 AM, Nick wrote: BTW, Its version 9.1.4 on a 2.2Ghz dual processor with 17GB of ram. Sample queries? EXPLAIN (ANALYZE, VERBOSE) ? How many concurrent queries? What's the connection limit on the pooler? How big is shared_buffers? -- Craig Ringer -- Sent via pgsql-general

Re: [GENERAL] Get DB Size by row

2012-07-21 Thread Bill Moran
On Fri, 20 Jul 2012 08:27:37 -0400 Benjamin Adams wrote: > Hello, > > I'm looking to do something like: > > select month, mb_size(description) from data where year = 2012 group by month > > I also am looking for raw size not compressed size. octet_length() will give you the size in bytes. I

Re: [GENERAL] Why is an ISO-8859-8 database allowing values not within that set?

2012-07-21 Thread Craig Ringer
On 07/21/2012 04:59 PM, Herouth Maoz wrote: Why is an ISO-8859-8 database allowing values not within that set? I am using Postgresql 8.3.14 on our reporting system. There are scripts that collect data from many databases across the firm into this database. Recently I added tables from a partic

Re: [GENERAL] A Better Way? (Multi-Left Join Lookup)

2012-07-21 Thread Tom Lane
"David Johnston" writes: > So, > EXPLAIN SELECT function_call(...) -- yields a planner expectation of 1 row > [Whereas] > EXPLAIN SELECT * FROM function_call(...) -- yields a planner expectation of > "result_rows" which defaults to 1000 Hm ... > Was this an intentional design decision to overr

Re: [GENERAL] A Better Way? (Multi-Left Join Lookup)

2012-07-21 Thread Tom Lane
I wrote: > "David Johnston" writes: >> So, >> EXPLAIN SELECT function_call(...) -- yields a planner expectation of 1 row >> [Whereas] >> EXPLAIN SELECT * FROM function_call(...) -- yields a planner expectation of >> "result_rows" which defaults to 1000 > Hm ... >> Was this an intentional desig

Re: [GENERAL] A Better Way? (Multi-Left Join Lookup)

2012-07-21 Thread David Johnston
> -Original Message- > From: pgsql-general-ow...@postgresql.org [mailto:pgsql-general- > ow...@postgresql.org] On Behalf Of Tom Lane > Sent: Saturday, July 21, 2012 5:48 PM > To: David Johnston > Cc: pgsql-general@postgresql.org > Subject: Re: [GENERAL] A Better Way? (Multi-Left Join Lookup

[GENERAL] postgresql ask for someparameters

2012-07-21 Thread leo xu
hello everyone: i found database crashed,i checked log,it indicated error: CONTEXT: xlog redo insert: rel 48777166/22362/48778276; tid 2/2 i wanted to know:what's meaning of real? and what's meaning of 48777166/22362/48778276 ?what's meaning of tid 2/2 ? please tell m