Re: [GENERAL] dblink to non postgresql dbms

2008-06-28 Thread David Rowley
Klint Gore wrote: > Maybe these things from pgfoundary might help > > http://pgfoundry.org/projects/dbi-link/ > > http://pgfoundry.org/projects/odbclink/ > > http://pgfoundry.org/projects/dblink-tds/ Thank you for the links. I'm just getting into looking at them now. I had forgotten to say be

Re: [GENERAL] query planner weirdness?

2008-06-28 Thread Steve Atkins
On Jun 28, 2008, at 12:53 AM, Bob Duffey wrote: If you're iterating through the records with a cursor, the plan may be different, IIRC - weighted to provide first row quickly, as opposed to the query that was run that's weighted to provide last row quickly. I agree, and I was hoping that would

Re: [GENERAL] query planner weirdness?

2008-06-28 Thread Tom Lane
"Bob Duffey" <[EMAIL PROTECTED]> writes: > 2008/6/28 Steve Atkins <[EMAIL PROTECTED]>: >> If you're iterating through the records with a cursor, the plan may >> be different, IIRC - weighted to provide first row quickly, as opposed >> to the query that was run that's weighted to provide last row qu

Re: [GENERAL] freebsd + postgresql 8.3.3 = 100% of cpu usage on stats collector?

2008-06-28 Thread hubert depesz lubaczewski
On Sat, Jun 28, 2008 at 02:22:26AM -0300, Rodrigo Gonzalez wrote: > So, read man page, and at least ktrace -C and read the output at > ktrace.out file. ok, i've read it and didn't understand. it says how to disable tracing but it doesn't say anything about enabling tracing. yesterday ktrace -p p

Re: [GENERAL] freebsd + postgresql 8.3.3 = 100% of cpu usage on stats collector?

2008-06-28 Thread hubert depesz lubaczewski
On Fri, Jun 27, 2008 at 06:04:19PM -0400, Tom Lane wrote: > Anything unusual about this DB (lots of tables, for instance)? sorry, missed that question - nothing unusual. couple of dbs (10-15), around 20 tables per databases, only 1 database really used - all others tend to be used only for testing

Re: [GENERAL] query planner weirdness?

2008-06-28 Thread Bob Duffey
2008/6/28 Steve Atkins <[EMAIL PROTECTED]>: > > On Jun 27, 2008, at 9:53 PM, Adam Rich wrote: > > >> >>> "Bob Duffey" <[EMAIL PROTECTED]> writes: >>> I'm seeing some query plans that I'm not expecting. The table in >>> question >>> is reasonably big (130,000,000 rows). The table h

Re: [GENERAL] ERROR: concurrent insert in progress

2008-06-28 Thread Ganbold
It shows following when I try to reindex: snort=# reindex index ip_src_idx; ERROR: concurrent insert in progress So I think it might be hard disk problem as Tom suggested. Unfortunately I don't have replacement hard disk now, so I guess I might change the DB location for a while. thanks, Ganbo